17 Technologies of People Tracking

Published by Ronny Max on

People Tracking Technologies generate actionable insights to increase conversation rates and profits of physical locations

Regardless of how you track people, the location position and time-based data quantify the In-Store Customer’s Journey in retail stores, shopping malls, airports, stadiums, and smart cities.

To put it simply –

By tracking people’s behaviors in your physical location, you build a data-driven decision process and get more conversions, sales, and profits from the assets you already have.


If you think you need in-store customer tracking, you do.

Technologies deployed to track people behaviors include:

  • (AI Deep Learning) Vision
  • (Biometrics) Facial Recognition
  • (Biometrics) Facial Demographics
  • (Biometrics) Eye-Tracking
  • 3D Spatial Learning (Augmented Reality)
  • 3D Stereo Video Analytics
  • 2D Monocular & Fisheye Video Analytics
  • Thermal Imaging
  • Time of Flight (ToF)
  • Structured Light 3D Scanner
  • Lidar 3D Laser Scanning
  • Open Source Raspberry Pi
  • WiFi (Wide Area Network) Location Tracking
  • UWB (Ultra Wide Band) Radar Imaging
  • BLE (Bluetooth Low Energy) Beacons
  • GPS (Global Positioning System) Personal Trackers
  • RFID (Radio Frequency Identification) Tags & Tracking

People Tracking Technologies quantify human behaviors by location, time, and activity.

The tracking solution can be interactive or anonymous. It could track objects, devices, or any “things” to capture the behaviors of a real person.

The complexity of the tracking system depends on location positioning, recognition attributes, and precision parameters.

Details matter.

Ronny’s note: I’m updating this post as the impact of the coronavirus is hitting United States of America. I live in the epicenter in Florida, and non-essential businesses and beaches closed. I often speak with people from China to Chile, in Europe, and around the world. The situation is dire and scary, primarily if your livelihood depends on physical retail. That said I would remind you that “people counting” become a viable market during the 2008 recession. The objective of “tracking people” in physical locations is to do more with what you already have. That is also true about caring for your family, your social circle, and your business. Stay safe.  

As a decision-maker, you should evaluate the tracking technology in context to your company’s technical infrastructure, analytics skills, and market positioning.

Before diving deeper into the tracking technologies, you should ask –

  1. Is your intent marketing to customers or manage the store?
  2. How do the people tracking data fit the client’s policies?
  3.  Which @Edge Sensors or Corporate/Cloud-Based Processing?

These questions are especially important in the evaluation of Deep Learning AI technologies on influencing in-store shopping behaviors.

The advanced technologies expanded the ability to quantify behaviors beyond location and time. It includes facial sentiments, body motions such as gestures, and customer engagement with products.

To sift between hype and value, evaluate the people tracking data by the quantity of training, the quality of the errors, and your business goals.

Most importantly,

Taken together, the massive shift to Cloud, 5G, and Internet of Things (IoT) ecosystems has sped up the ability of retailers, malls, and other physical hubs to better track, understand and manage people in real life.

In-Store Optimization (customer analytics techniques) | Behavior Analytics Academy
In-Store Optimization (ISO) | Behavior Analytics Academy

3 Traps for In-Store Customer Tracking

Many retailers, brands, malls, agencies, and others, suffer from three common traps when evaluating people tracking solutions.

The traps come from a lack of understanding of what is the value of the tracking technology, and how the solution achieves a specific objective within the context of the client’s organization.

Trap #1: Customers or Physical Store?

The first trap is the focus on consumer-facing solutions instead of stores. When thinking about people tracking, many clients think in terms of “smart mirrors” or “marketing to customers inside the store”.

Retail Store or Customers? | Behavior Analytics Academy

As you see below,

Location Marketing is about interactions with potential customers. But Anonymous Location Analytics is about doing more from the available assets inside the store.

Trap #2: Focus on Cost instead of Profit

For example, if the conversation is about adding staff at the fitting rooms, many retailers think in terms of the additional payroll.

But fitting rooms are high up in the in-store purchase funnel for retailers who sell apparel. By adding staff that helps and encourages customers to use the fitting rooms, the retailer will see more conversations and more sales.

The profit analysis, therefore, takes into account the added payroll costs, but also the increase in sales due to the higher conversations in the fitting rooms.

Analyze the complete picture.

Trap #3: Building “Widgets” not “Solution”

Sadly, the “widget” trap is a mutual reinforcement challenge for both clients and providers. It comes from the “silver bullet syndrome”, the magical “widget” that will solve all problems.

Here’s my take –

True and effective innovation comes from building a process for innovation.

The strength of an organization comes from the internal ability to process knowledge and utilize the information in skills and activities.

People Tracking Solutions are about empowering the client’s business users (such as store managers, corporate professionals, and executives) to do a better job.

Practice an open data policy.

Open data policy by Ulta Beauty (Source Ronny Max)

Geo-Location Analytics (Spatial Intelligence)

Location intelligence, or spatial intelligence, is the process of deriving meaningful insight from geospatial data relationships to solve a particular problem. It involves layering multiple data sets spatially and/or chronologically, for easy reference on a map, and its applications span industries, categories, and organizations.

Wikipedia: Location intelligence

In retail, malls, and real-estate property management, Location Analytics measures the value of the physical location and the bricks-and-mortar store opportunity based on demographic and competitive data.

Location Analytics is also used inside the physical store to evaluate product positioning and customer engagement.

Here are common business benefits from Location Analytics:

  • Shopper Traffic to store (#Visitors)
  • Proximity Traffic (% Capture Rate)
  • Choosing a Physical Store Site (Trade Area Analytics)
  • InStore Product Positioning (Path to Purchase)
  • InStore Employee Locations (Service Productivity KPIs)
  • InStore Live Map & Product Information (Path Analysis)

50-60% of perceived value is captured from location-based data.

McKinsey Global Institute

Anonymous or Personal Location Analytics

“Even at the highest levels, retailers don’t really understand either the rewards or the risks associated with anonymous vs. non-anonymous customer geo-location data.”

RSR Research, Location Analytics: New Data, New Opportunities

Remember that –

There is a difference between the solutions that identify shoppers by name (Identity) and those that track people anonymously (random ID).

Most importantly,

Interactive technologies require the customer’s consent.

The customer’s opt-in is done online in a digital form. It can happen in the operating system (Google or Apple), application (Facebook or Shopify), or the telecom companies.

As a result, interactive technologies need to adhere to consumer privacy standards.


Since the customer’s consent often belongs to a third party, retailers and brands are now stepping up on the business model of Direct-to-Consumer.

Interactive tracking is also an integral feature of digital devices for employees. Most store systems – from scheduling to task management – also generate location and time data.

Anonymous tracking refers to systems where captured objects are recognized by attributes (such as the head of a person) and identified by a random ID.

The point is –

Tracking people means “quantify behaviors.”

The tracking technologies track objects or signals. The people tracking solution can be Location Marketing (with Consumer Opt-In) or Location Analytics (Anonymous).

Location-Based Marketing

The objective of location-based marketing is as simple as it sounds:

Track the location of, and connect with, the personal device of a person for the business objective of marketing.

The challenge in location marketing is attribution. It means the solution needs to identify not only the location of the device but also the device itself and its owner, and across every step in the purchase funnel.

That’s why location marketing is digital marketing based on location.

For example, Google “Near Me” is an example of real-time customer tracking.

Even if the objective is real-time marketing, there are challenges to retailers and brands:

  • The consumer’s consent is often to a Third Party such as Facebook and Google.
  • Young adults tend to have more than one device (attribution)
  • Older adults may not have smartphones (no GPS/WiFi Tracking)
  • Some people deride real-time location tracking due to privacy concerns (me!)
  • Store performance requires data on all behaviors (where anonymous tracking excels)

In physical locations, in-store location marketing is consumer-facing and interactive technologies such as Smart Mirrors and Digital Fitting Rooms, QR product information, and self-service kiosks.

Another solution where location marketing is useful is in the building of buyer profiles, especially those with high Customer Lifetime Value (CLV) that are active in the loyalty programs.

Here’s the thing –

Store Performance metrics don’t need the shoppers’ identity. On the contrary, for malls and retailers, in-store analytics work better with anonymous shopper tracking.

The first generation of anonymous tracking started with “People Counting.”

Shopper Tracking: Beyond People Counters

People counters are sensors or devices that “count” the number of people that pass a physical or virtual line. The technologies “detect” the object (or signal) and generate a linear dataset.

The frequent use of “people counting” is in retail stores and shopping centers where people become “visitors” once they cross the threshold and enter the physical location. With foot traffic, you can evaluate and compare the store opportunity.

Some solutions go beyond “Door Counting” and provide metrics such as Sales Conversation and the real-time applications of Queue Management and Service Intensity.

Technology has a huge impact on the nature of the data, for example:

  • People Counting identifies “heads”
  • WiFi tracks the location of Smartphones
  • BLE Beacons uses opt-in customer data
  • RFID tags and tracks product SKUs
  • Facial Demographics identifies gender
  • Biometrics captures individual faces
  • Vision Analytics can “recognize” images

The ingenuity of the solution providers takes many forms. People Counting is used to control energy usage. Queue management concepts used for real-time services. And startups are amazingly creative in building solutions for In-Store Optimization.

For example, here’s an innovative concept by Intelligenxia:

Smart Bathroom Cleaning: By tracking the location of employees and how long it takes to clean each bathroom, the property owner saw an increase of 58% efficiency.

The people tracking solution includes 3D sensors at the entrance of the bathroom that count the number of people entering the bathroom. Once the threshold of 18 people passed, the system triggers an alert to a smartwatch application. When the cleaner enters the bathroom, the solution registers the time of arrival and departure after cleaning. Activity-Based Costing provided the analytics framework designed by Intelligenxia. 

Smart Bathroom Cleaning by Intelligenxia
Smart Bathroom Cleaning by Intelligenxia

One more topic of confusion to clarify 🙂

@Edge Sensors vs. People Trackers

There are two ways to measure shopping behaviors in a physical location: foot traffic sensors and device-based tracking.

Footfall Sensors with Edge Computing:

Foot traffic sensors vary in their ability to process Edge Computing. It means the person tracker captures the raw data and process the metrics inside the sensor. The output is a data file that includes the location and timestamp of the objects.  

In sensors, we care about the accuracy of tracking ALL objects within the Field of View.

Standard traffic sensors have people tracking technologies such as 3D Video Analytics, Thermal Imagining, and Time of Flight.

Device-Based Personal Trackers:

In tracking people, device-based solutions make the underlying assumption that the device represents the behavior of a specific individual.

That is the distinct difference between sensors and devices.

While sensors generate information on a Field of View, the wireless signal from the device provides data on the path of a single person.

The wireless technologies are distinct by their range and the accuracy between “captured” and “actual” position. In that sense, the technology tracks the behavior of an individual device.

GPS, WiFi, and BLE Beacons are examples of device-based customer tracking technologies.

This is important –

All People Tracking Technologies have five core attributes: Detection, Recognition, Prediction, Precision, and Time-Based Precision.

Since each tracking technology has benefits and challenges, in-store analytics solutions are often a complementary mix of technologies.

People Tracking Technologies (2020)

Ronny’s disclaimer: Behavior Analytics Academy and Silicon Waves are technology and solution provider agnostic. The following is not a recommendation.

(AI Deep Learning) Vision

30% of retailers will have computer vision technology in place over the next 12 months

RIS Retail Technology Study: Retail Accelerates (2019)

Vision Analytics works by recognizing patterns in images. The Deep Learning AI software translates the images to data, context, and action.

The output is a quantified description of the image, such as Stanford University’s cat or a person. Vision technology has many applications, from driverless cars to medical imagining.

In retail, Vision technology works hand-in-hand with facial recognition and product imaging in these three applications:

  • Automated Stores: Vision technology is used to identify the people who walk into the store (facial recognition), grab a product (product imagining), and walk out (connect identity to checkout process). The most known example is Amazon Go.
  • Inventory Visibility: Vision technology is embedded in robotics. The robots stroll the shelve and search for out-of-stock and assortment compliance events.
  • In-Store Customer Engagement: Vision technology is found in beauty applications, personalized promotions, and recognizing VIP customers as they enter the store.

Vision is a set of Deep Learning AI algorithms. The expertise of the solution provider comes from the process of training the software to solve a specific problem in retail or other physical locations.

Deep Learning AI Vision application in Retail | Behavior Analytics Academy

In people tracking technology,

Vision technology powers up In-Store Customer Tracking.

You can imagine this vast potential with in-store analytics startups such as Modcam (Sweden), Perceive (US), and Aura Vision Labs (UK),

(Biometrics) Facial Recognition

Facebook biometric software works as well as the human brain. Facial Recognition is no longer a question of technology.

Retailers, brands, and shopping centers need to consider two factors when dealing with facial recognition: government regulation and consumer sensitivity.

There are significant differences between continents, countries, and communities in attitudes to privacy and regulation.

The European Union created GDPR. In China, the government actively supports the development of facial recognition technology.

In the United States, the retailers and brands are conflicted between the obvious benefits of facial recognition and concerns about a backlash due to inaccuracy and privacy.

To deal with consumer sensitivity to privacy, you have 3 options:

  • Facial Identification: the technology compares an image to a given image within a database. Facial Identification is often used for security and surveillance.
  • Facial Selfies: the same tech as facial identification with one caveat – the person uploads self-image, tags it, and consent to sharing the data. Used in loyalty programs.
  • Facial Demographics: the software process facial features and the output is data on gender and age. The identity of the face remains anonymous.

The nuances of consumer privacy and consent are muddled by HOW and WHERE the data is stored and processed.

(Biometrics) Facial Demographics

In 2020, Facial Demographics became a viable technology in camera-based tracking solutions used for In-Store Analytics.

The difference between facial recognition and facial demographics is in the importance of attributes such as the shape of the eye, and chic to chin ratio.

Biometrics attributes provide demographic and behavioral information on visitors to physical stores without requiring to knowing their identity.

Regardless of the anonymity of facial attributes, solution providers have the ability to learn about a wide variety of behaviors, for example:

  • Gender
  • Returning visitors
  • Staff recognition
  • Age groups
  • Skin Attributes
Facial Attributes Example in Saphora Studio (Source Ronny Max)

(Biometrics) Eye Tracking

Eye Tracking measures the relative motion of the eye to the position of the head.

In online Conversation Rate Optimization (CRO), eye-tracking technology offers valuable insights to what people see on the webpage.

The use of eye-tracking technology for physical retail is new.

The tracking technology is a combination of training the software to recognize images (Deep Learning AI) and the context of a physical location, such as a window, or storefront, or a display.

Eye Tracking Technology, for example, is used to analyze the Visibility Rate in product positioning and customer engagement studies.

3D Spatial Learning (Augmented Reality)

Augmented Reality “adds” images, sounds, and text, to what we see in the real world. Powerful analytics engines drive the personalization of the software. Tracking shopping behaviors is a side-affect.

The craze over Augmented Reality started with Pokemon Go. In retail, the applications are both online, for example, the Gap Dressing Rooms and offline, for example, Sephora Beauty Studio.

A fascinating new technology is the combination of Vision, Biometrics, and Augmented Reality in Sentiment Analysis.  Watch for Microsoft.

3D Stereo Video Analytics

Stereo video sensors are designed for tracking objects across the camera’s Field of View.

The stereo sensors include a high-resolution camera and processor for the three-dimensional capture of the object. The 3D architecture compensates for occlusion and shadows by adding depth (distance from the camera).

The high level of tracking accuracy allows for tracking more complex behaviors such as high-volume traffic, queue management, and customer engagement.

Premium 3D Stereo sensors go beyond people counting and path analytics, and also provide information on gender recognition, group counts, gaze direction, and stuff exclusions.

Solution providers include TD Intelligence, Brickstream (Flir), Eurecam, Hella, and Xovis.

Live demo in NRF 2020 by Xovis

2D Monocular & Fisheye Video Analytics

Monocular sensors capture images through a single-lens camera. It also includes the fisheye lens, which is an ultra-wide-angle lens for wide panoramic images.

Regardless of the lens, the sensor process the image, and the output is data counts.

For in-door people counts, monocular devices can achieve 90% accuracy in 90% of stores.

This is a cost-effective tracking solution for mall stores and in-store events.

The fisheye cameras (used primarily for security surveillance) and monocular sensors are widespread.

Video analytics is found in smart cameras of Axis, Bosch, and Panasonic.

Thermal Imaging

Thermal Imaging detects emissions from moving objects. Since thermal technology is not sensitive to light, it can function in any physical space. The challenge is the “blending” of a person’s heat signature with the surrounding environment.

Thermal sensors are easy to install and calibrate, with common 95% counting accuracy for stores located in shopping centers. Solution providers include Delopt and Irisys.

Time of Flight (ToF)

Time of Flight detects the time of light between the camera and the object. By sending the laser beams to many directions, the sensor knows the exact positioning of objects.

BEA Helma for example, embedded people counting software directly into door sensors.

Structured Light 3D Scanner

Structured Light projects a known pattern, or stripes, on a three-dimensional object. The array of lights strikes the surface and calculates the depth and surface of objects.

3D Scanners offer a low-cost option for people counters.

Lidar (3D Laser Scanning)

Lidar deploys scanning technology that sends laser and measures the return times and wavelengths to create a three-dimension visualization of the targeted object.

Lidar is an acronym standing for light imaging, detection, and ranging.

The Lidar technology is often used in mapping and survey application and is recently being used in retail and smart buildings.

Open-Source Raspberry Pi

Raspberry PI is open-source hardware that serves as an alternative to proprietary people counting systems.

A tracking solution that contains open-source hardware, open-source software such as Linux, and off-the-shelf 3D camera, is a low-cost technology for people counting and tracking.

A packaged solution offered by specialized middleware integrators, such as Enliteon, solves the challenges of accuracy and support.

GPS Personal Tracker

Global Positioning System (GPS) is a network of orbiting satellites. A GPS tracking system uses the microwave signals between the Global Navigation Satellite System (GNSS) and GPS devices to track the location, speed, time, and direction of an object.

GPS is built-in the Apple and Android operating platforms, and therefore the tracking technology provides real-time and historical data on the customer’s journey.

During the years 2015 to 2017, Google’s location-based queries based on Near Me grew 900% (see above). Today, retailers, mall owners, and other physical locations can subscribe to Google Store Visits.

WiFi Location Tracking

WiFi is a standard for Wireless Local Area Network (WLAN). Antennas capture the radio waves from mobile phones and cover a range of up to 100,000 square feet.

MAC Address is a unique identifier per device, and therefore we can assume it represents an individual customer. The data output depends on the customer’s phone and the active activation of the features of WiFi.

WiFi technology gained ground in people tracking applications because it is relatively easy to deploy and is cost-effective, especially for large venues such as shopping centers and stadiums. Most importantly, most people use the technology to access emails and social media on the go.

In addition to big providers such as Cisco’s Meraki, many solution providers use WiFi tracking in their technology basket.  

BLE Beacons

Bluetooth Low Energy (BLE) Beacons function in the wireless range between NFC and WiFi.

Beacons are the first-mover technology for tracking (opt-in) loyalty customers in physical stores. The rise of loyalty applications led to an increase in the deployment of Beacons technology.

Today, the widespread of GPS/WiFi technologies are improving fast to work indoors. At the same time, lower costs and infrastructure requirements are pushing the market for other alternatives.

Beacons platforms are Google’s Eddystone and Apple’s iBeacon

Ultra-Wide-Band (UWB) Radar Imaging

Ultra Wide Band (UWB) functions in low-energy, short-range, and high-bandwidth environments. UWB comes from Radar Imaging and is entering the retail space.

UWB core advantage is accuracy, which can be 5 to 10 cm. The low power usage and location positioning accuracy are big advantages for In-Store Analytics.

RFID Location Tracking

RFID (Radio Frequency Identification) uses electromagnetic fields to track the tags attached to objects.

In retail, RFID is primarily used for supply chain and loss prevention applications by tagging and tracking the movements of “things.” As RFID adoption rate gains momentum, retailers are deploying the technology to enhance the in-store customer experience.  

Tracking Tree: What’s the BEST Tracking Technology?

Here’s how I evaluate an in-store technology strategy QUICKLY.

The Tracking Tree is a series of questions that retailers should ask, and solution providers should answer.

Let’s dive in.

Step #1: Identify Business Objectives

If your client is a retailer, a brand, or a marketing agency, identifying the client’s business objective is the first step:

  • If you want to send push notifications to customers, you need an opt-in and opt-out features and process
  • If you want to know the store’s sales opportunity, you need an accurate measure of Visitors
  • If you want to evaluate the performance of mall stores, you need proximity traffic and capture rate
  • If you want to improve the customer’s experience in the checkout, you need Queue Management
  • If you want to optimize Product Positioning, you need to test for Optimal Engage Time

Since every tracking technology has pros and cons, by pinpointing the primary objective to the decision-maker, you are now in the position to evaluate the technology.

Step #2: Identify Tracking Technology

Regardless of the client and provider, don’t assume people know how to evaluate in-store tracking solutions. Here are topics to discuss:

  • Location-Based Marketing or Location Analytics
  • Sensors or Device-Based Tracking
  • People Counting (Detection) or Tracking Technologies
  • User Experience & Analytics Interface
  • Metrics Framework & Key Performance Indicators (KPI)

Now that you know the business objective and the position of the tracking technology within the stack, it is time to close the sale.

Step #3: Identify Solution Provider

Clarity Creates Confidence.

  • The scope of the Solution Provider
  • The Pricing Structure
  • The Accuracy Framework
  • Get the details on Analytics
  • Professional Services

Here’s the answer.

The BEST In-Store Customer Tracking Technology is the BEST solution for the business objectives of the Client

That’s it.

The Tracking Tree method works exceptionally well for today’s disruptive market where the technologies can change overnight

Which Tracking Technology Company?

In addition to the tracking technology, geographical locations, professional services, and business models differentiate between solution providers.

Specifically, there are categories of players:

Category #1: People Counting (Footfall Tracking) Companies that ventured into In-Store Analytics

People Counting companies and Value Added Resellers are natural first movers. In addition to the specialized companies mentioned above, here are additional providers (in no particular order):

Value Added Retailers (VARs) may mingle a variety of technologies. AirBits from Argentina uses WiFi and BLE Beacons. InteliPower from South Africa and Australia combine WiFi Tracking and 3D sensors, as well as experimenting with new technologies Lidar, eye-tracking, and AI Vision.

Anonymous Path Analytics (live demo by Prodco)

Category #2: Retail Enterprise Providers go up the technology stack

Enterprise network security and deployment companies are providing installment, calibration, and professional services for people tracking solutions, including:

Mall Counts (Source Ash Projects)

Category #3: Marketing expertise turns toward physical retail

We see data-driven marketing-minded startups such as InReality in Atlanta, Nurama in Brussels, and Digital Mortar in San Francisco.

The trend of startups with digital marketing and software-as-service expertise is strong today and is changing the pace of In-Store Customer Tracking.

Category #4: Location-Based Companies go big into retail

Another facet of this trend is Geo-Spatial Intelligence companies who are focusing on retail, for example, Esri and Orbital Insights.

That trend is a reflection of how location-based data and demand analytics are valuable beyond retail, for example, food scarcity.

NYC Uber Engineering

Category #5: Internet of Things (IoT) Companies become platforms

Significant software and consulting companies are entering the market with end-to-end retail analytics platforms. IBM and Intel, for example, are creating ecosystems of connected solutions and advanced analytics as part of their Internet of Things (IoT) offerings.

Capgemini, for example, includes in-store tracking in its retail practice.

While the impact of the coronavirus is still unclear, the physical store still provides retailers, brands, mall management, and other location owners, the direct touch with their customers.

In-Store Technologies are crossing the chasm.

People Tracking technologies are becoming more cost-effective and more accurate in real-time.

Tier-1 companies such as Amazon, Alibaba, and Walmart, are leading the trend to seamless retail. And consumer expectations are accelerating.

Moreover, new data-oriented professions are emerging. With “data-driven conversions” every part of store operations, marketing, sales, profits, and growth gets easier

Physical Retail is NOT a Black Box

Click here to subscribe

Ronny Max

Ronny Max is an author, speaker, and executive coach. In 2015, Ronny served as the domain expert at Stanford University Vision Project, the first research venture into AI technologies for physical retail. In 2017, she founded the Behavior Analytics Academy. Ronny Max coined and developed the In-Store Optimization (ISO) frameworks for physical environments.