Location measurement is more than just knowing a latitude and longitude point. In fact, assigning the closest place to a latitude and longitude point results in an incorrect match of place to a user’s location in more than 90 percent of cases – just one of the findings revealed in our latest white paper, Measuring Offline Consumer Behavior: Understanding the Foundation of Location Measurement and Analytics.
This report takes a look at the science, challenges and opportunities behind location analytics and what this emerging field means for companies seeking customer insights and competitive intelligence from location data. Here’s a look at five key takeaways from the report:
- Location + Place = Foundation of Analytics: Location measurement involves two equally important and complex components: location data collection and accurate place assignment.
- “Smart” Data Collection: Cell tower signal, Wi-Fi and GPS are the primary methods of collecting location data via smartphones. GPS provides the most accurate location measurement with an average accuracy range of 5 to 30 meters. Sensor data, such as a smartphone’s accelerometer, compass and gyroscope, are another important set of factors used to determine a user’s movement in the physical world.
- Optimized for Battery Life: Although GPS provides the most accurate location data, it is the most battery intensive and thus has the most potential to negatively affect user experience. Placed has devised intelligent algorithms optimized for battery life, utilizing sensor requests to determine the best time to collect high-quality data. In an experiment, using optimized algorithms resulted in battery drain of 2% per hour compared to 11% drain when not optimized.
- Meta Makes the Model: Place databases are noisy and studies have shown this noise makes a significant impact on accuracy. In 9 of 10 cases, assigning the closest place to a latitude and longitude point results is the wrong match of place to a user’s actual location when depending solely on database information. The Placed inference model leverages metadata, such as time, demographic affinity, business popularity, name normalization and business category, to significantly improve assignment accuracy and yield more actionable analytics for clients.
- Privacy Focused, Privacy Forward: Best practices for location data collection are transparent and keep consumers’ privacy at the core of their methodology. Placed has pioneered an explicit, triple opt-in approach to location measurement that beats industry standards.
To read the full report, please visit: Measuring Offline Consumer Behavior: Understanding the Foundation of Location Measurement and Analytics.