Retail Site Selection Data: Why the Street Matters More Than the Spreadsheet
I've watched enough retail chains expand and contract to know the pattern. The demographics look perfect. The foot traffic model sings. The competitor analysis is bulletproof. And eighteen months later, the store is closing. What happened?
What happened is that nobody looked at the actual street. The spreadsheet treated the location as a dot on a map. The customer experienced it as a cracked pavement, a burned-out streetlight, and a puddle they had to step around to get to the door. Data can't smell neglect. People can.
What the spreadsheet misses
Standard site selection models are brilliant at counting people. They're terrible at describing the experience of being one of those people. Here's what I've seen kill stores that looked perfect on paper:
Accessibility. Is the entrance visible from the street? I've seen stores tucked behind pillars, down steps with no handrail, past a loading zone that blocks the sidewalk half the day. The customer has to want to find you. Most don't.
Surroundings. What is the condition of neighbouring properties? I stood outside a flagship candidate in Stockholm where the demographics were gold-plated. The neighbouring building had peeling paint, boarded windows, and a smell I couldn't identify. Customers notice. They vote with their feet.
Maintenance burden. A location with deteriorating infrastructure — broken lighting, damaged pavements, overgrown vegetation — doesn't just look bad. It costs you. Either you absorb the hit to your brand, or you pay to fix public assets that aren't yours. Neither shows up in the rent calculation.
Competitor adjacency, physical. A competitor across the street may be invisible in census data but highly visible to a pedestrian. I've seen a bakery open opposite an established café because the spreadsheet said "low competitor density." On the ground, the café had a queue out the door and the bakery lasted eight months.
The physical condition score
Landvex adds a physical condition dimension to location data. For any candidate site, we score what the customer actually experiences:
Street condition. Pavement quality, lighting, cleanliness, signage legibility. The first impression before they reach your door.
Access quality. Ease of approach by foot, bicycle, car, public transport. Obstacles. Barriers. The friction between "I want to go there" and "I give up."
Neighbourhood maintenance. The condition of surrounding properties, public assets, green space. A proxy for whether the area is loved or abandoned. Customers sense this before they can name it.
Change trajectory. Is the area improving, stable, or declining? A location with moderate demographics but a rising trajectory can outperform a "prime" spot that's sliding. The spreadsheet won't catch the slide until the rents drop.
From score to decision
Strong demographics with poor physical condition is a trap. The customers exist, but the experience of reaching you suppresses conversion. I've seen stores in theoretically perfect locations struggle because the approach felt unsafe, inconvenient, or simply depressing.
Moderate demographics with improving physical condition? That's where the smart money goes. The spreadsheet undervalues it. The competitors miss it. You get in before the rent catches up with the reality.
The physical condition score doesn't replace demographic analysis. It completes it. It adds the dimension that is observable, comparable, predictive — and almost never included.
Portfolio-level application
For retailers with multiple locations, physical condition data changes everything. Which stores are in declining areas that need attention or exit? Which are in improving areas where additional investment pays off? Where should the next store go, based on both customer potential and street-level experience?
I've seen portfolio strategies flip when this data comes in. Stores that looked similar on paper turned out to be in radically different physical environments. The ones in declining streets needed different marketing, different staffing, different capital plans. Or they needed to not exist.
Request a pilot →
The spreadsheet tells you where the people are. The street tells you whether they'll actually walk through your door.