Measuring Offline Consumer Behavior:
Measuring Offline Consumer Behavior: Understanding the Foundation of Location Measurement and Analytics
January 2013
For more information, please contact: Placed, Inc. [email protected]
Executive Summary: Location analytics is emerging as a rapidly growing area of consumer market research. Location analytics provides insights into offline consumer behavior and contextualizes mobile usage to help bridge the gap between digital and physical world activities. The foundation of location analytics lies in sound location measurement.
This report addresses the challenges and potential opportunities in location measurement, highlighting several takeaways that will be important for any company that is seeking to understand the emerging world of location analytics, the measurement that supports it, and how to utilize these insights to support their business. Below is a summary of the key areas discussed in this report:
Location measurement involves two equally important and complex components: data collection and place assignment.
The primary methods of collecting location data via smartphones include cell tower signal, Wi-Fi and GPS. Among these three methods, GPS provides the most accurate location measurement with an average accuracy range from 5 to 30 meters.
Sensor data is another important piece of information that can be obtained via smartphones and used to generate accurate location analytics. The types of sensors that provide useful data include the accelerometer, compass and gyroscope.
Although GPS provides the most accurate location data, it is the most battery intensive and has the most potential to negatively affect user experience if collection methods are not optimized to intelligently maximize battery life.
The limitations of place databases, including out-of-date information, incomplete business information, etc., make assigning a specific place to a user’s location challenging. In early experiments, Placed found that assigning the closest place to a latitude and longitude point resulted in an inaccurate match of place to location more than 90 percent of the time.
In order to address the limitations in place databases, multiple sources must be taken into account to reach optimal accuracy in assigning a place to a location. Placed’s inference model addresses this issue by leveraging over two million points of ground truth.
Collecting any form of consumer information requires the utmost transparency and respect for consumer privacy. Best practices for data collection respect consumers’ privacy at the core of their methodology, which includes explicit opt-ins for location measurement.
2
Table of Contents
Introduction ………………………………………………………………………………………………… 4 What is Location Analytics? ……………………………………………………………………………… 4 Current Methods and Challenges in Location Data Collection ………………………………………. 5 Cell Tower Signal ………………………………………………………………………………… 5 Wi-Fi ………………………………………………………………………………………………. 6 GPS ……………………………………………………………………………………………….. 7 Sensor Data Value ………………………………………………………………………………. 8 Battery Drain Challenges ……………………………………………………………………….. 8 The Noise in Location Data: Place Assignment Accuracy……………………………………………. 9 Factors that Influence Place Assignment ……………………………………………………... 9 Inference Modeling for Place ……………………………………………………...……………. 10 Privacy Principles ……………………………………………………...…………………………………. 11 Conclusion ……………………………………………………...…………………………………………. 11
3
Introduction Today, more than 90 percent of retail sales occur offline. Yet offline consumer measurement tools have significantly lagged behind online methods, giving online retailers a significant advantage in capitalizing on consumer behavior. However, this situation is poised to change with the rise of location analytics. Location analytics is quickly becoming an important component for planning, implementing and measuring both offline and online strategies, not only for retailers but for companies across industries. According to ABI Research, a market intelligence company, the location analytics market is expected to reach $9 billion by 2016 as location becomes a valuable and indispensable metric by which marketers plan, execute and evaluate strategies.
In this new era of location analytics, brands, advertisers and publishers need dependable knowledge of offline consumer behavior in order to understand this emerging space and the potential benefits to their businesses. Those businesses that recognize location measurement is significantly more complex than a latitude and longitude point will be well positioned to gain actionable insights into their customers, as well as their competitors, from this fast-growing segment of analytics.
What is Location Analytics? It is important to define location analytics as referenced in this report. Location analytics is the contextualization of location data into actionable insights that reveal behavioral patterns of consumer activity in the physical world. Location analytics is often confused with location-based advertising. The two are not synonymous, although location analytics can be used to build more intelligent location-based advertising strategies.
Location measurement is the method by which data is collected in order to derive analytics. In order to produce high-quality insights, measurement practices must be sound.
There are two elements to location measurement that form its foundation: data collection and place assignment. This report will delve into each of these topics in order to provide a better understanding of location measurement and analytics, providing an overview of methodology, challenges and best practices.
4
Current Methods and Challenges in Location Data Collection Today’s advancements in cell tower signal strength, Wi-Fi and a growing number of GPS-enabled phones would lead many to assume that collecting location information via smartphones is a simple process that leads to highly accurate results. Although these technologies continue to advance, accurate location measurement via these methods faces challenges that are unique to each approach.
Cell Tower Signal: Location Data Collection Method Cell tower data is collected when users connect to their cellular service provider. Location data collected via cell tower provides the least accurate view into the location of the user. Cell tower triangulation location accuracy ranges from 600 to 3,000 meters.
Below is a depiction that demonstrates the accuracy of location data derived from cell tower data. The star icon represents the cell phone user’s actual location. The blue circle represents the location range of the user as reported by the cell tower signal. In this case, the cell tower accuracy is 1,014 meters (0.6 mile). The circle's perimeter shows the range of possible locations given the low accuracy of this method.
Range of user location identified by Cell Tower Signal
Location data results via cell tower collection method. Accuracy = 1,014 meters. Source: Placed, Inc.
5
Wi-Fi: Location Data Collection Method Location data collected via Wi-Fi provides a more accurate view of location data than cell tower data, but also represents challenges in accuracy. Location data collected via Wi-Fi ranges from 25 to 150 meters in accuracy.
Below is a depiction that demonstrates the accuracy of location data from Wi-Fi. The star icon represents the cell phone user’s actual location. The blue circle represents the location range of the user as reported by Wi-Fi data. In this case, Wi-Fi has an accuracy of 114 meters. The circle's perimeter shows the range of possible locations given the low accuracy. Although significantly improved compared to cell tower data collection, Wi-Fi still leaves much error in determining location metrics. People can also disable Wi-Fi on their devices, making data collection via this method not possible in those cases.
Range of user location identified by Wi-Fi
Location data results via Wi-Fi collection method. Accuracy = 114 meters. Source: Placed, Inc.
6
GPS: Location Data Collection Method Among the location measurement methods covered in this report, GPS provides the most accurate location data available today. GPS location data is collected via GPS-enabled devices, with an accuracy that ranges from 5 to 30 meters.
Below is a depiction that demonstrates the accuracy of location data collected via GPS. The star icon represents the place the user is actually visiting. The small circle represents the location of the user as reported by GPS data. In this case, GPS provides an accuracy reading of 7 meters. Further, in this example the user could be walking towards or away from the parking lot of the business given the GPS range. Notice that the circle’s perimeter has significantly decreased in this example compared to cell tower and Wi-Fi, demonstrating significantly improved accuracy in location data collection.
Range of user location identified by GPS
Location data results via GPS collection method. Accuracy = 7 meters. Source: Placed, Inc.
7
Although GPS data collection provides the most accurate representation of a user’s location, this method faces several challenges as well. As with Wi-Fi data collection, GPS must be enabled on phones to utilize this method of data collection. This means that when users disable GPS on their devices, location measurement via this method is not possible. Secondly, collecting GPS data is the most battery intensive of the three methods. Battery drain represents a significant challenge that requires balance between gathering location information without negatively impacting user experience.
Sensor Data Value Sensor data provides information about the mobile device’s physical movement, including orientation and acceleration. Placed uses sensor data to understand a user’s state of motion; whether they are stationary, in transit or walking. In conjunction with GPS data, this helps Placed better understand user activity as well as provides a signal for when a user might be in motion. Knowing when a user is in motion helps optimize data collection to limit battery impact.
Battery Drain Challenges Battery drain due to GPS data collection depends heavily on the user’s device and operating system. The most important factor that will influence battery drain will be the device hardware. Logically, older, cheaper phones tend to lose battery more quickly, while newer, higher-end devices tend to have stronger battery power. The proliferation and fragmentation of devices also poses a challenge in predicting the effects of battery drain.
Placed conducted an experiment on GPS data collection and battery drain. The experiment results showed that when location data collection was not optimized for battery life, the battery drain on the phone was 11 percent per hour in tests. This level of battery drain would significantly impact the user experience and is therefore not an ideal method of data collection.
In order to account for the inherent battery drain associated with GPS usage, Placed has devised algorithms that are optimized for battery life. These algorithms intelligently utilize sensor requests to determine the best time to start collecting high-quality data. This method maximizes the quantity and quality of location data with constraints on battery drain. Utilizing battery-optimization algorithms resulted in a battery drain of 2 percent per hour during this experiment, significantly improved from the 11 percent experienced in the non-optimized experiment.
8
The Noise in Location Data: Place Assignment Accuracy Collecting location data is the first challenge in location measurement and analytics. The second, and equally important component, is determining the actual place to assign to that location point and user. In the depiction below, we find dozens of possible places that are located within 200 meters of the person, identified here as the red icon.
Place possibilities within 200 meters. Source: Placed, Inc.
Factors that Influence Place Assignment Inaccuracies in place databases significantly hinder accurate place assignment. Examples of place databases include government sources, user-generated sources, and privately licensed databases, among others. Unfortunately, these databases have limitations. These limitations include incorrect addresses of locations, lack of uniformity in naming conventions (i.e. Starbucks, Starbucks on Second, Starbucks Coffee, etc.) and out of date data that doesn’t reflect the present landscape of establishments that are currently operational.
9
In early experiments, Placed found that 90 percent of the time, assigning the closest place to a latitude and longitude point resulted in an incorrect match of place to location. This is especially problematic in highly concentrated areas such as shopping centers and urban pockets where accurate place assignment can be a challenge. Placed addresses this challenge by leveraging its inference modeling to significantly increase accuracy in place assignment, and therefore derive significantly more accurate and actionable insights from location data.
Inference Modeling for Place In order to account for the limitations in place databases, Placed has created models to intelligently infer place in order to create more accurate location analytics. Given the proprietary nature of this scientific modeling, the below provides only a partial overview of Placed’s inference modeling.
One of the methods utilized by the Placed inference model is the use of metadata in order to contextualize physical-world visits. This approach is based on the philosophy that the value of a place is much more than its latitude and longitude; these are the details that make location analytics actionable for companies. Below provides examples of several metadata components used:
Time: Time of day and hours of the establishment are utilized to help determine the likelihood of a visit occurring between places. For instance, if GPS identifies a person as standing on the corner of the street equidistant from two establishments there is no way to determine which of the two establishments the user is most likely to be visiting. But if we take into account that the hour of the visit is 1:00 a.m. and the two neighboring businesses are Macy’s and a bar, then intelligent modeling will help infer the visit was to the bar rather than to Macy’s.
Popularity: Business popularity in a given geographic area can also be utilized to help assign a visit to a place given the probability of visitation by the overall number of visits occurring to a place.
Demographic Affinity: Overlaying demographic data collected via panelists helps build further inference based on demographic affinity. For instance, women would have a higher affinity for visiting stores that cater to women’s needs versus men’s needs and therefore can be used as a component in inferring place.
Naming Conventions: Given that the same business location could have more than one name in some databases and therefore be classified as two separate business entities, such as “Starbucks on nd
2 Ave.” and “Starbucks on 2
nd
and Seneca,” the Placed inference modeling utilizes a deduplication
algorithm to solve this issue.
Category of Business: Business category features are also used in Placed’s inference model. For example, if a user stayed at a place for two hours this user is more likely to be at a movie theater rather than the supermarket next door.
10
Privacy Principles Collecting any form of user information requires the utmost transparency and respect for consumer privacy. The measurement industry across a broad range of mediums – including TV, radio, online and mobile – has adopted various methods to collect consumer activity and data including surveys, cookies and opt-in panels. Best practices for data collection keep respect for consumers’ privacy and information at the core of any methodology.
Given the sensitivity of location data collection, Placed believes that a transparent, opt-in method of data collection provides the most privacy-conscious approach to location analytics. Unlike the standard in cookiebased web measurement (which requires consumers to opt-out), Placed requires users to explicitly opt-in before collecting any location data. Further steps to ensure user privacy include the aggregation of location data across users for reporting purposes. Aggregation of data ensures that an individual’s location is not revealed in the analytics generated by Placed.
Conclusion Location analytics is emerging as a critically important area in market research, comparable to other analytics areas including TV, web and mobile measurement. In order for location analytics to reach its full potential in delivering actionable insights, methodology approaches must address both the data collection and place assignment components of location measurement.
As we have seen, both data collection and place assignment present challenges that require significant research and scientific modeling in order to optimize for collection and accurately assign the proper place to a user’s location. Only after these two criteria are met can location analytics reach a level of accuracy that provides actionable insights for businesses.
As a pioneer in location analytics, Placed has collected, analyzed and validated billions of location data points to develop the most accurate and complete location analytics available, supported by the world’s largest location database. This location data opens the door to a new era in offline analytics that provides transparency into the patterns of consumer activity in the physical world. With this intelligence, companies can better understand trends within their industries, gauge performance against competitors and look for new opportunities to leverage shifts in consumer behavior to drive growth in today’s highly complex and competitive market.
As location analytics becomes a common set of metrics for businesses in planning, executing and evaluating their strategies, Placed has released several best practice case studies to help businesses navigate this emerging landscape. To learn more about these real-life use cases for location analytics, please visit: https://www.placed.com/resources/case-studies-and-white-papers
11
About Placed, Inc. Placed, Inc. is the leader in location analytics. By connecting the physical and digital world, Placed quantifies consumers’ real-world behaviors into actionable insights, arming offline and online companies with powerful data to drive their business and stay ahead of competition. Founded in January 2011, Placed is headquartered in Seattle and is backed by Madrona Venture Group. For more information about Placed and its location analytics solutions, please visit: https://www.placed.com/
For more information, please contact: Placed, Inc. [email protected]
12
January 2013
For more information, please contact: Placed, Inc. [email protected]
Executive Summary: Location analytics is emerging as a rapidly growing area of consumer market research. Location analytics provides insights into offline consumer behavior and contextualizes mobile usage to help bridge the gap between digital and physical world activities. The foundation of location analytics lies in sound location measurement.
This report addresses the challenges and potential opportunities in location measurement, highlighting several takeaways that will be important for any company that is seeking to understand the emerging world of location analytics, the measurement that supports it, and how to utilize these insights to support their business. Below is a summary of the key areas discussed in this report:
Location measurement involves two equally important and complex components: data collection and place assignment.
The primary methods of collecting location data via smartphones include cell tower signal, Wi-Fi and GPS. Among these three methods, GPS provides the most accurate location measurement with an average accuracy range from 5 to 30 meters.
Sensor data is another important piece of information that can be obtained via smartphones and used to generate accurate location analytics. The types of sensors that provide useful data include the accelerometer, compass and gyroscope.
Although GPS provides the most accurate location data, it is the most battery intensive and has the most potential to negatively affect user experience if collection methods are not optimized to intelligently maximize battery life.
The limitations of place databases, including out-of-date information, incomplete business information, etc., make assigning a specific place to a user’s location challenging. In early experiments, Placed found that assigning the closest place to a latitude and longitude point resulted in an inaccurate match of place to location more than 90 percent of the time.
In order to address the limitations in place databases, multiple sources must be taken into account to reach optimal accuracy in assigning a place to a location. Placed’s inference model addresses this issue by leveraging over two million points of ground truth.
Collecting any form of consumer information requires the utmost transparency and respect for consumer privacy. Best practices for data collection respect consumers’ privacy at the core of their methodology, which includes explicit opt-ins for location measurement.
2
Table of Contents
Introduction ………………………………………………………………………………………………… 4 What is Location Analytics? ……………………………………………………………………………… 4 Current Methods and Challenges in Location Data Collection ………………………………………. 5 Cell Tower Signal ………………………………………………………………………………… 5 Wi-Fi ………………………………………………………………………………………………. 6 GPS ……………………………………………………………………………………………….. 7 Sensor Data Value ………………………………………………………………………………. 8 Battery Drain Challenges ……………………………………………………………………….. 8 The Noise in Location Data: Place Assignment Accuracy……………………………………………. 9 Factors that Influence Place Assignment ……………………………………………………... 9 Inference Modeling for Place ……………………………………………………...……………. 10 Privacy Principles ……………………………………………………...…………………………………. 11 Conclusion ……………………………………………………...…………………………………………. 11
3
Introduction Today, more than 90 percent of retail sales occur offline. Yet offline consumer measurement tools have significantly lagged behind online methods, giving online retailers a significant advantage in capitalizing on consumer behavior. However, this situation is poised to change with the rise of location analytics. Location analytics is quickly becoming an important component for planning, implementing and measuring both offline and online strategies, not only for retailers but for companies across industries. According to ABI Research, a market intelligence company, the location analytics market is expected to reach $9 billion by 2016 as location becomes a valuable and indispensable metric by which marketers plan, execute and evaluate strategies.
In this new era of location analytics, brands, advertisers and publishers need dependable knowledge of offline consumer behavior in order to understand this emerging space and the potential benefits to their businesses. Those businesses that recognize location measurement is significantly more complex than a latitude and longitude point will be well positioned to gain actionable insights into their customers, as well as their competitors, from this fast-growing segment of analytics.
What is Location Analytics? It is important to define location analytics as referenced in this report. Location analytics is the contextualization of location data into actionable insights that reveal behavioral patterns of consumer activity in the physical world. Location analytics is often confused with location-based advertising. The two are not synonymous, although location analytics can be used to build more intelligent location-based advertising strategies.
Location measurement is the method by which data is collected in order to derive analytics. In order to produce high-quality insights, measurement practices must be sound.
There are two elements to location measurement that form its foundation: data collection and place assignment. This report will delve into each of these topics in order to provide a better understanding of location measurement and analytics, providing an overview of methodology, challenges and best practices.
4
Current Methods and Challenges in Location Data Collection Today’s advancements in cell tower signal strength, Wi-Fi and a growing number of GPS-enabled phones would lead many to assume that collecting location information via smartphones is a simple process that leads to highly accurate results. Although these technologies continue to advance, accurate location measurement via these methods faces challenges that are unique to each approach.
Cell Tower Signal: Location Data Collection Method Cell tower data is collected when users connect to their cellular service provider. Location data collected via cell tower provides the least accurate view into the location of the user. Cell tower triangulation location accuracy ranges from 600 to 3,000 meters.
Below is a depiction that demonstrates the accuracy of location data derived from cell tower data. The star icon represents the cell phone user’s actual location. The blue circle represents the location range of the user as reported by the cell tower signal. In this case, the cell tower accuracy is 1,014 meters (0.6 mile). The circle's perimeter shows the range of possible locations given the low accuracy of this method.
Range of user location identified by Cell Tower Signal
Location data results via cell tower collection method. Accuracy = 1,014 meters. Source: Placed, Inc.
5
Wi-Fi: Location Data Collection Method Location data collected via Wi-Fi provides a more accurate view of location data than cell tower data, but also represents challenges in accuracy. Location data collected via Wi-Fi ranges from 25 to 150 meters in accuracy.
Below is a depiction that demonstrates the accuracy of location data from Wi-Fi. The star icon represents the cell phone user’s actual location. The blue circle represents the location range of the user as reported by Wi-Fi data. In this case, Wi-Fi has an accuracy of 114 meters. The circle's perimeter shows the range of possible locations given the low accuracy. Although significantly improved compared to cell tower data collection, Wi-Fi still leaves much error in determining location metrics. People can also disable Wi-Fi on their devices, making data collection via this method not possible in those cases.
Range of user location identified by Wi-Fi
Location data results via Wi-Fi collection method. Accuracy = 114 meters. Source: Placed, Inc.
6
GPS: Location Data Collection Method Among the location measurement methods covered in this report, GPS provides the most accurate location data available today. GPS location data is collected via GPS-enabled devices, with an accuracy that ranges from 5 to 30 meters.
Below is a depiction that demonstrates the accuracy of location data collected via GPS. The star icon represents the place the user is actually visiting. The small circle represents the location of the user as reported by GPS data. In this case, GPS provides an accuracy reading of 7 meters. Further, in this example the user could be walking towards or away from the parking lot of the business given the GPS range. Notice that the circle’s perimeter has significantly decreased in this example compared to cell tower and Wi-Fi, demonstrating significantly improved accuracy in location data collection.
Range of user location identified by GPS
Location data results via GPS collection method. Accuracy = 7 meters. Source: Placed, Inc.
7
Although GPS data collection provides the most accurate representation of a user’s location, this method faces several challenges as well. As with Wi-Fi data collection, GPS must be enabled on phones to utilize this method of data collection. This means that when users disable GPS on their devices, location measurement via this method is not possible. Secondly, collecting GPS data is the most battery intensive of the three methods. Battery drain represents a significant challenge that requires balance between gathering location information without negatively impacting user experience.
Sensor Data Value Sensor data provides information about the mobile device’s physical movement, including orientation and acceleration. Placed uses sensor data to understand a user’s state of motion; whether they are stationary, in transit or walking. In conjunction with GPS data, this helps Placed better understand user activity as well as provides a signal for when a user might be in motion. Knowing when a user is in motion helps optimize data collection to limit battery impact.
Battery Drain Challenges Battery drain due to GPS data collection depends heavily on the user’s device and operating system. The most important factor that will influence battery drain will be the device hardware. Logically, older, cheaper phones tend to lose battery more quickly, while newer, higher-end devices tend to have stronger battery power. The proliferation and fragmentation of devices also poses a challenge in predicting the effects of battery drain.
Placed conducted an experiment on GPS data collection and battery drain. The experiment results showed that when location data collection was not optimized for battery life, the battery drain on the phone was 11 percent per hour in tests. This level of battery drain would significantly impact the user experience and is therefore not an ideal method of data collection.
In order to account for the inherent battery drain associated with GPS usage, Placed has devised algorithms that are optimized for battery life. These algorithms intelligently utilize sensor requests to determine the best time to start collecting high-quality data. This method maximizes the quantity and quality of location data with constraints on battery drain. Utilizing battery-optimization algorithms resulted in a battery drain of 2 percent per hour during this experiment, significantly improved from the 11 percent experienced in the non-optimized experiment.
8
The Noise in Location Data: Place Assignment Accuracy Collecting location data is the first challenge in location measurement and analytics. The second, and equally important component, is determining the actual place to assign to that location point and user. In the depiction below, we find dozens of possible places that are located within 200 meters of the person, identified here as the red icon.
Place possibilities within 200 meters. Source: Placed, Inc.
Factors that Influence Place Assignment Inaccuracies in place databases significantly hinder accurate place assignment. Examples of place databases include government sources, user-generated sources, and privately licensed databases, among others. Unfortunately, these databases have limitations. These limitations include incorrect addresses of locations, lack of uniformity in naming conventions (i.e. Starbucks, Starbucks on Second, Starbucks Coffee, etc.) and out of date data that doesn’t reflect the present landscape of establishments that are currently operational.
9
In early experiments, Placed found that 90 percent of the time, assigning the closest place to a latitude and longitude point resulted in an incorrect match of place to location. This is especially problematic in highly concentrated areas such as shopping centers and urban pockets where accurate place assignment can be a challenge. Placed addresses this challenge by leveraging its inference modeling to significantly increase accuracy in place assignment, and therefore derive significantly more accurate and actionable insights from location data.
Inference Modeling for Place In order to account for the limitations in place databases, Placed has created models to intelligently infer place in order to create more accurate location analytics. Given the proprietary nature of this scientific modeling, the below provides only a partial overview of Placed’s inference modeling.
One of the methods utilized by the Placed inference model is the use of metadata in order to contextualize physical-world visits. This approach is based on the philosophy that the value of a place is much more than its latitude and longitude; these are the details that make location analytics actionable for companies. Below provides examples of several metadata components used:
Time: Time of day and hours of the establishment are utilized to help determine the likelihood of a visit occurring between places. For instance, if GPS identifies a person as standing on the corner of the street equidistant from two establishments there is no way to determine which of the two establishments the user is most likely to be visiting. But if we take into account that the hour of the visit is 1:00 a.m. and the two neighboring businesses are Macy’s and a bar, then intelligent modeling will help infer the visit was to the bar rather than to Macy’s.
Popularity: Business popularity in a given geographic area can also be utilized to help assign a visit to a place given the probability of visitation by the overall number of visits occurring to a place.
Demographic Affinity: Overlaying demographic data collected via panelists helps build further inference based on demographic affinity. For instance, women would have a higher affinity for visiting stores that cater to women’s needs versus men’s needs and therefore can be used as a component in inferring place.
Naming Conventions: Given that the same business location could have more than one name in some databases and therefore be classified as two separate business entities, such as “Starbucks on nd
2 Ave.” and “Starbucks on 2
nd
and Seneca,” the Placed inference modeling utilizes a deduplication
algorithm to solve this issue.
Category of Business: Business category features are also used in Placed’s inference model. For example, if a user stayed at a place for two hours this user is more likely to be at a movie theater rather than the supermarket next door.
10
Privacy Principles Collecting any form of user information requires the utmost transparency and respect for consumer privacy. The measurement industry across a broad range of mediums – including TV, radio, online and mobile – has adopted various methods to collect consumer activity and data including surveys, cookies and opt-in panels. Best practices for data collection keep respect for consumers’ privacy and information at the core of any methodology.
Given the sensitivity of location data collection, Placed believes that a transparent, opt-in method of data collection provides the most privacy-conscious approach to location analytics. Unlike the standard in cookiebased web measurement (which requires consumers to opt-out), Placed requires users to explicitly opt-in before collecting any location data. Further steps to ensure user privacy include the aggregation of location data across users for reporting purposes. Aggregation of data ensures that an individual’s location is not revealed in the analytics generated by Placed.
Conclusion Location analytics is emerging as a critically important area in market research, comparable to other analytics areas including TV, web and mobile measurement. In order for location analytics to reach its full potential in delivering actionable insights, methodology approaches must address both the data collection and place assignment components of location measurement.
As we have seen, both data collection and place assignment present challenges that require significant research and scientific modeling in order to optimize for collection and accurately assign the proper place to a user’s location. Only after these two criteria are met can location analytics reach a level of accuracy that provides actionable insights for businesses.
As a pioneer in location analytics, Placed has collected, analyzed and validated billions of location data points to develop the most accurate and complete location analytics available, supported by the world’s largest location database. This location data opens the door to a new era in offline analytics that provides transparency into the patterns of consumer activity in the physical world. With this intelligence, companies can better understand trends within their industries, gauge performance against competitors and look for new opportunities to leverage shifts in consumer behavior to drive growth in today’s highly complex and competitive market.
As location analytics becomes a common set of metrics for businesses in planning, executing and evaluating their strategies, Placed has released several best practice case studies to help businesses navigate this emerging landscape. To learn more about these real-life use cases for location analytics, please visit: https://www.placed.com/resources/case-studies-and-white-papers
11
About Placed, Inc. Placed, Inc. is the leader in location analytics. By connecting the physical and digital world, Placed quantifies consumers’ real-world behaviors into actionable insights, arming offline and online companies with powerful data to drive their business and stay ahead of competition. Founded in January 2011, Placed is headquartered in Seattle and is backed by Madrona Venture Group. For more information about Placed and its location analytics solutions, please visit: https://www.placed.com/
For more information, please contact: Placed, Inc. [email protected]
12
