How to steal your competitors' real estate strategies using public location data
All you need is a list of their locations
The tl;dr
Analyzing competitors’ location data can be a powerful guide and benchmark for your real estate strategy
But getting your hands on location data can be time consuming and, if done manually, prone to errors
Data providers like ChainXY can help you unlock the power of location data
This type of analysis can help you understand your footprint potential or saturation, cannibalization risks, and even outlet strategies
“Fish where the fish are”
Every time I’ve built out a real estate strategy for a retailer, I always include a competitor benchmarking analysis. Understanding where your competitors and peer brands are is an important input to determine market potential and site selection. Smaller brands should generally “fish where the fish are” (ie where your competitors have already generated demand and conditioned consumers to shop at location XYZ), and in some cases take advantage of the market research that larger brands have already done.
As an example, you can probably rest assured that Nordstrom did a fair amount of market research to validate that people who like higher priced clothing probably live in the area where their stores are (although they flopped when it came to entering a new country).
This certainly isn’t true all the time (not all of Nordstrom’s locations achieved expectations, and surprisingly not all big box brands actually even do that much research when selecting new store locations), but looking at benchmarks can still help you answer three important questions as you expand your store fleet:
Size of the prize: How many stores should we have (across the country, and within specific markets)?
Market prioritization and site selection: Where should they be?
Format selection: What type(s) of stores should they be?
If nothing else, you’d better be equipped with a response when someone asks you “how many stores does our competitor have?”
And whether the competitor’s real estate decision was a good one or not, it doesn’t change the fact that they’ve conditioned some set of customers to shop there for that product. Might as well take advantage of the work they did.
As one of my former boss’s always quipped: “pioneers get the arrows, settlers get the gold.”
Benchmarking requires complete and clean data
The general approach to competitor analysis can be broken down into the following steps:
Identify your competitor(s): whose market share are you taking from? where are your customers cross shopping?
Build a list of their locations, along with addresses and any other info you deem necessary (eg lat-long coordinates)
Augment the data with other location attributes (eg sole store in market, one of X stores in market, etc.)
Overlay the data with your own first party data (customers, sales in market, etc)
Visualize the outputs
Determine the “so what” based on the output
So with that, here’s a case study using a real brand’s (anonymized) data:
Let’s start by pretending your customers also shop Nordstrom
…so, you believe their footprint is a good benchmark for you (which is probably applicable to most DTC brands). While competitor benchmarking should include at least a few other companies, we’re sticking with just one for simplicity.
So we’ve identified our competitor (step 1 of 6).
Now we need to build a list of their locations. You can generally do this 1 of 3 ways:
Extract it from their Locations page: This is manual and, in some cases like Nordstrom, actually not possible if their page is just a map with a query function. Especially if you’re dealing with a benchmark that has hundreds of locations to copy and paste, this is a terrible use of anyone’s time.
Phone a friend: Some brokers, landlords, or even operators at the brand may have a location list handy, but that’s a tough ask in some cases. “Can you send me your company’s full list of locations, broken down by type (eg full price vs outlet), and with complete addresses?” And even if you do get it from someone, it can be tough to trust that it’s clean of typos, and more importantly up-to-date.
Buy it from ChainXY: You can buy a full list of Nordstrom’s locations for <$100, AND feel good about it being complete, accurate, and clean. I’ve personally used their data a number of times, and it’s always worth it.

For the most basic analysis, I only need three columns of data:

That said, ChainXY data also comes with a number of other data points for each location that you might find helpful as you’re analyzing competitors’ footprints and thinking about your own:
Location features: important as you contemplate your own locations’ additional features
Drive Through
ATMs
Delivery
Store Hours: important as you assess your operating hours and labor models
Location first appeared: useful for understanding when a competitor opened in the market
Lat / long coordinates: useful for mapping in a GIS platform
And many others
Step 2 (make a locations list) is complete.
Now we need to augment our data with attributes we want
This will vary from brand to brand, so you should start by first identifying which questions you’re trying to answer. Here are my usual go-tos:
How many stores does this competitor have in total, and in each CBSA? This will help me quantify the “size of the prize”
How many markets have they doubled down in? What’s the maximum amount of stores this brand has put into a single market? This helps me understand cannibalization risk, as well as the opportunity size for each individual market. More stores = higher likelihood for more sales and customers (but not always).
How does their store footprint compare to my existing customer base? This helps me understand the existing overlap in market concentration between my brand and their brand, which could help me (a) identify market opportunities (where my customer base is small but they have a larger store base), and (b) how good of a benchmark this competitor really is (ideally our top markets are the same).
How do they think about the intersection of their full price and outlet strategies? To the extent you may be thinking about introducing an outlet model, a natural question will be: “how do others plan their outlet footprints given the existing full price footprint?”
To answer all of these questions, you’ll need to augment the competitor location data with just a couple things, and then do some minor summary extracts:
You can use a zip code to MSA converter to tag these locations with their respective markets.
And then build a nice summary table like the below:
Step 3 is now complete (augment the data with other location attributes).
Now it’s your turn
The Market Summary view is where we can finally start to see some interesting things in the raw data. In the snippet above, for example, we can already see (1) major concentrations in the broader LA metro for full price and, perhaps more interestingly, (2) that their Rack locations are often 2-3x the amount of Full Price locations — in the same market.
But I digress…
Now it’s your turn: add in your own data. I used a real example of a $100m brand for this:
This will likely require some collaborative work with your data or finance team: a lot of brands don’t have CBSAs in their data warehouse, but you can download a Zip Code x MSA lookup table for free (or close to free) online, and append your sales data with it to achieve this output.
For simplicity, we’re only going to look at sales and pretend you don’t have existing stores. But if you have stores or other channels, then you should absolutely add these to your market tables to visualize any comparisons.
Step 4 is now complete.
And now for the fun parts
The data is cleaned, uploaded, and sorted. Now to visualize and summarize.
As a starting point, I always like to compare my markets’ composition to that of my benchmarks. Here’s what that’d look like:

In this case, it’s actually nice to see that your curve follows a similar direction as that of Nordstrom: this indicates there’s a strong correlation between your customer base and their store footprint. The opposite case would be if the red line wasn’t as smooth. The reason this curve sits below (but essentially parallel to) Nordstrom, is because Nordstrom doesn’t have sales in all 393 MSAs, but this brand does (so the “long tail” is pulling the whole curve down).
This is just one of many ways to visualize a competitor’s footprint data. It might also be helpful to visualize it in this way too:

Graphics are pretty, but now we need to extract something from them
I’m a big fan of looking at subtotals and intersections when it comes to data. And in this example, remember, we want to understand the following:
How many stores does this competitor have in total, and in each CBSA?
Answer in table belowHow many markets have they doubled down in? What’s the maximum amount of stores this brand has put into a single market?
Answer in table belowHow does their store footprint compare to my existing customer base?
See above chartsHow do they think about the intersection of their full price and outlet strategies?
Answer can be deduced from the table below
Interestingly, Nordstrom only has 1 market that is solely a Full Price store: Bridgeport-Stamford-Norwalk CT which makes sense given the demographics (very affluent). And while many markets have both Full Price and Rack, there are 46 markets that are Rack only — these are unsurprisingly the markets that tend to be #51-100 for most brands.
These data points and visuals would lead me to wonder:
Rack proximity to Full Price stores might be critical for inventory access: excess inventory from their Full Price goes to Rack, and pretty easily with this footprint strategy (as a reaction to the 42 overlapping markets)
The markets that serve Full Price customers are the same markets that serve Rack customers (as a reaction to the 42 overlapping markets). We can’t conclude whether it’s the same customer between the two though, because larger CBSAs also just have larger populations.
They may be geographically isolating their Rack locations to protect the brand (as a reaction to the 46 Rack only markets)
The Rack model may be the only way to break into smaller markets (as a reaction to the 46 Rack only markets)
We can’t say any of the above with certainty, but it gives me some great ideas for any brand that is wondering what an outlet strategy could look like.
And all because of the location data from ChainXY.
Be aware of the limitations!
Competitor benchmarking is a great way to get one of many inputs into your strategic development processes. But you should be aware of the dangers of using competitors for benchmarking in this particular way:
Just because they did it, doesn’t mean you should too: often times you’ll hear a landlord or broker tell you “well your competitor/peer is at this location so you should be too.” Keep in mind, America is notoriously overstored; brands have more square footage in this country vs anywhere else, and are also very frequently closing stores as a result.
Location data doesn’t tell you much, if anything, about timing or deal terms:
Deal terms: a brand may be in a particular location because the deal was really favorable. This happens more than you’d expect.
Timing: a brand may be successful in a location, but it may have taken them years to get there. A store that’s been in a market for 10 years will of course be more successful than a store that’s been there for just 1. And just because the store is now successful, doesn’t mean it was a great financial decision.
New developments: using benchmarks of existing stores means you’re closing off the possibility of thinking about new developments, where stores (and perhaps the entire development itself) don’t yet exist.