Bringing Online Analytics To The World of Bricks and Mortar
By Dan Gildoni
While I’m guessing you’ve heard the term location intelligence before, and have at least a vague understanding of the concept, let’s first address a popular misconception. Location intelligence is not about the location. It is about the people coming to and interacting with the location. Because without the interaction of humans, a location is meaningless (at least from a business perspective).
Looking to understand the performance of specific locations is nothing new, particularly for the real estate, retail and city management sectors. But while data is being used to inform decision making, it is generally static and very broad data, such as census data highlighting a population increase in a specific city or neighbourhood, or real estate price trends. Data regarding ‘who’ is visiting a location has not been available, meaning many decisions are based on incomplete, one-dimensional and possibly out-dated information.
In Search of Data Insights
Technology has been trying to solve this and improve location-based decision-making by providing precise data to inform location-based decisions. This has mainly involved electronic devices, known as people counters, used to measure the number of visitors entering and leaving a location. These devices, in general, can be classified as laser, wi-fi, and video systems. But the data they produce is limited. They only capture a small snapshot of the total footfall picture and sacrifice a broader, more nuanced understanding of who is actually visiting a location. More detailed information, such as how long visitors spend in a place, where they come from, what route they take and their associated demographic profile, and so on, is well beyond their limited capabilities.
The result is that those looking to get reliable insights into bricks and mortar locations, unlike their online counterparts, are left relying on very imprecise information to make decisions that annually amount to billions and billions of Euros.
Big Data Footfall Analysis
However, the technology is available to finally perfect footfall analysis and bring deeper insights. Artificial intelligence, machine learning and big data analytics are being combined to generate precise and deeper data regarding the footfall performance of any location. This finally equips the traditional world of bricks and mortar with tools previously only available in the online world. This is transforming location-based decision-making in the real estate, retail, outdoor advertising, city planning sectors and beyond.
The gateway to this information is the smartphone, given just about everyone happens to have one in their pocket. But the key that unlocks this data is the mobile app.
Mobile apps provide access to billions of GPS data points. These can be used to understand the movement of people and their interactions with the physical world. Now I know it sounds kind of creepy, but don’t worry, it’s not. The reason we’re able to use this data is because we have solved the privacy issue. We use only anonymized data by applying a proprietary process to cleanse it, so personal information cannot be jeopardized. This fully anonymized data is also aggregated and, of course, GDPR compliant. Importantly, data is only sourced from mobile apps that require consent and detail how anonymized data will be used and shared.
Bridging the data gap between what’s available to online retailers and regular real-world stores would not be possible without GPS. Its geographical precision (5 to 10 meters) is far greater than cellular triangulation (around 200 meters).
Nevertheless, GPS data has its challenges. For example, there may be different mobile app usage patterns across different population groups. This means the user base of a specific app may not be representative of the population as a whole. Different mobile apps can also have varying usage patterns across different activities (e.g. dining vs shopping). These issues will skew the data if not dealt with properly.
In order to resolve these inconsistencies, and project realistic footfall metrics for a given location, PlaceSense uses a patent-pending proprietary machine learning algorithm. This allows us to calibrate the data using machine learning, which consists of multiple mathematical models continuously trained and tested in various regions/locations. This allows us to make sense out of data, gathered from millions of devices, by understanding the aggregated groups of visitors to any location. This provides our customers with the ability, for example, to understand the various catchment areas where visitors to a location came from, where they visited before, and where they go afterwards.
Billions of Data Points
PlaceSense relies on large data panels of millions of unique devices to achieve this. These devices generate billions of monthly pings, which when filtered and enriched by our algorithms gives a highly-accurate representation of reality. We then cross reference the results with additional data sources to understand things like social demographics, purchasing power, gender and age. This information brings immense value to decision makers, especially to those looking to get tenants for retail rental units or retailers pursuing an expansion strategy. For the first time they can understand not only the numbers of people visiting a location, but the profiles of the various types of visitors.
With access to the latest in AI, big data and advanced analytics, the world of location-based information is being transformed. Retailers can now optimize their businesses to attract more consumers. Retail estate professionals have visibility into how and how often people travel to and from for their shopping centers, and those of their competitors. They can view and analyze cross visitation and loyalty patterns of visitors between various locations. Importantly, property owners can also benchmark their location against others to gauge its health and potential.
These AI-powered data-driven insights are finally placing analytics at the heart of real-world location-based decision making. Businesses are using these insights to analyze people’s interactions with their stores and those of their competitors, for example, and to inform and back their decisions.
Large-scale, real-world data is at last being democratized, beyond the needs of the advertising industry, and is finally giving access to the types of hard facts people in the real world need to help make businesses and cities flourish.
About Dan Gildoni
Dan is the CEO and Co-founder of PlaceSense A serial entrepreneur and business leader with over 15 years of experience in B2B and B2C digital platforms, Dan has had years of experience leading a diverse range of online businesses that brought him into PropTech.