Why Retail Property Needs a Location-AI Disruptor
- Matt Copus
- Nov 18
- 5 min read
Stop Paying Monopoly Prices for Blunt Retail Analytics
Retail property analytics hasn’t kept up with the market it’s meant to serve.
If you’re a landlord, investor, retailer or leasing agent, you probably know the current options:
Premium banking-data products with genuinely strong signal, but heavy price tags and usage restrictions.
“Trade area” analysis where someone draws a neat shape on a map and calls it insight.
Cheap data feeds and self-serve products that look impressive in a pitch deck but fall apart when you try to make a real investment decision.
None of that is good enough as your main decision engine for multi-million-dollar assets.
Loculyze exists because we were tired of watching serious decisions made on rough guesses, partial coverage and sales theatre.

Banking data: strong signal, narrow beam
Let’s be clear: banking data is good data.
It’s transaction-level, it’s audited, and it’s built on real spend. For the assets it covers, in the geographies it covers, it can provide powerful insight.
The issue isn’t the quality. The issues are:
Cost. These products are priced for very large portfolios and institutions. For everyone else, access is patchy, out-of-reach or non-existent.
Restrictions. Coverage is defined by the provider, not by your strategy. You can’t freely explore every centre, every competitor or every development scenario.
Control. If you depend on one exclusive data source, your view of the market is effectively leased, not owned.
So banking data is excellent where it’s available and justified. It’s just not a complete operating system for location intelligence.
That’s the gap we target.
The fallback: humans drawing shapes
When banking data is too expensive or unavailable, the market often falls back on “best guess” trade area analysis.
Someone draws a boundary on a map. Then they add up population, spend and demographics inside the line.
This method was already outdated 25 years ago. It assumes people behave neatly and predictably. They don’t.
As our technical co-founder likes to say: never send a human to do a machine’s job. So, if you can use data science to measure who actually visits a place, why would you bet capital on someone’s opinion of who might?
The other extreme: Dog Whistle Analytics
On the opposite end you have low-cost data and self-serve providers.
They sell scale: billions of data points, thousands of segments, endless dashboards.
What they usually don’t sell is answers.
We call this Dog Whistle Analytics:
Throw enough data points at the client and hope they hear something useful.
Many of these platforms come from ad-tech. They’re built to sell media, not to support underwriting, asset strategy or network planning.
Signal quality, sample bias and model performance are secondary. That’s fine if you’re optimising click-through rates. It’s risky if you’re backing a development, acquiring a centre or resetting a leasing strategy.
Why we built Loculyze
We started Loculyze with a simple view:
The industry doesn’t need more data. It needs better truth.
Open-source datasets and mobile signal data have become incredibly rich. Used properly, they can give you a live, unrestricted and cost-effective view of how customers interact with every location—not just the ones a bank chooses to cover.
That matters because it unlocks:
Landlords growing asset value with precise, evidence-based strategies.
Investors spotting under-performing centres and high-potential sites before the market prices them in.
Retailers getting an independent view of reach, performance and brand strength—not just landlord-supplied numbers.
Leasing agents targeting the right tenants with evidence, not anecdotes.
But getting there isn’t as simple as buying raw feeds.
Open and mobile data is messy. It has gaps, bias and noise. To be investment-grade it needs to be normalised, enriched and modelled by people who understand retail property, not just SQL.
That’s what Loculyze does.
LoculAI: our Location-AI engine
At the core of Loculyze is loculAI—our Location-AI platform.
Think of it as a pipeline:
Organise – Open Source Data: We start with privacy-safe mobile signal data, national statistics, open maps, POI databases and client data where available.We clean, align and standardise everything so it speaks the same language.
Enrich – locul Destinations, Occupiers, Customer, Spend, Client Data: We build a detailed view of:
Destinations and centres (what’s there, who operates it, what’s around it).
Occupiers and categories (from anchors to specialty).
Customer profiles and spend behaviour.
Any first-party data you provide (sales, leasing, transaction records).
Predict – loculChoices is our spatial interaction model.It estimates the probability that a customer will visit each destination, given all the choices available to them—not just where they live, but how they actually move and shop.
Measure – loculWays turns mobile signals into clear, live metrics:
Who visits.
Where they come from.
How often they show up.
How your asset is performing relative to its potential and to competitors.
The result: location intelligence you can put in front of an investment committee without crossing your fingers.
How Loculyze is built to be different
We made a few hard calls early that still guide everything we do:
Customer-true, not postcode-true. We show who actually visits and buys, not just who lives near a centre.
No asset restrictions. If it’s on the map, we can analyse it. On-portfolio, off-portfolio, hypothetical. No need for landlord permission or hardware on site.
Open and affordable inputs. We use non-exclusive datasets so no one can hold your portfolio hostage.
Prescriptive, not descriptive. Our job is to turn data into strategies you can execute—to show you what could be and map the path to getting there.
Rigour over theatre. Our models are tuned against real performance data. If the numbers don’t line up, we fix the model—we don’t spin the story.
Sector expertise built-in. We’re commercial property, retail and analytics people first. The platform reflects how deals are actually done, not how a generic data platform thinks they should be.
Product first, consultancy where it counts. LoculAI is a platform, not a one-off report. You can self-serve through online patffoms , plug our outputs into your own tools, and lean on us when you need deeper strategic support.
We sit in a different spot on the spectrum: high-tech, customer-centric insight in the spirit of banking-data products, but with wider coverage, fewer restrictions and a very different cost base.
What this looks like in practice
Here’s how our clients typically use Loculyze:
A landlord benchmarks every centre in the portfolio, sees where they are under-penetrated in key catchments, and prioritises capex and remix strategies backed by evidence.
An investor screens a long-list of potential acquisitions, identifies centres that over-trade their catchment (and those that don’t), and bids with a clearer view of upside and risk.
A retailer maps true brand reach across a city, spots white-space locations, and goes into lease negotiations with their own independent numbers.
A leasing team matches listings with the tenants whose customers already move through that trade area, then builds a targeted, data-backed pitch.
No monopoly contracts. No mystery algorithms. No dashboards that look clever and say nothing.
Just a grounded view of how people use places—and what you can do about it.
If you’re still relying on guesswork, you’re paying twice
You pay once for the report, the licence, or the data feed.
You pay again when the asset under-performs because the inputs weren’t good enough—or weren’t available where you needed them.
Loculyze is built to stop that.
Banking data will remain an important part of the market. We’re not trying to replace it. We’re building the open, flexible, high-coverage layer that lets you operate with confidence everywhere else.
If you want to replace opinion and theatre with location intelligence that stands up to a red-team review, LoculAI is built for you.
Let’s talk about the assets you care about—and what the data really says about them.



Comments