I’m excited to share that we updated StreetLight InSight® again this week – and it’s an update that I’ve been eagerly anticipating for quite a long time. Beginning now, our clients can access Metrics derived from our new Location-Based Services data source directly from our StreetLight InSight web app – that’s our one-stop, cloud-based platform for the best Big Data resources and the processing software that makes them useful. So, why am I so excited about this? It means some of our most popular Metrics are even more comprehensive and accurate than before. That’s because our device sample size now represents about 10% of the U.S. population. We’re processing roughly 60 billion location data points into travel pattern analytics every month – and counting!
Defining Location-Based Services Data
The Location-Based Services data now available in StreetLight InSight are derived from smart phones with apps that use Location-Based Services. Think of weather, retail shopping, or dating apps, all of which provide services to their users that are fundamentally linked to those users’ locations. See Figure 1 below for an example of Location-Based Services in action.
Figure 1: Location-Based Services provide specific utilities to smartphone users that are based on their location. Users know when their location is being tracked when a small, solid arrow icon appears in the right-hand corner of their smart phone’s display - look where the orange arrow is pointing in the image above.
We are now using this data source instead of cellular tower data for the majority of our Site Metrics, which describe the travel patterns and characteristics of visitors to specific locations, or “Sites.” (Longtime users know that we used to call these our “Retail Metrics.” We’ve changed the name because transportation experts have found them useful, too.) The main advantages of Location-Based Services data over cellular tower data are:
- Spatial precision: On average, our Location-Based Services data has 25 meter spatial precision or better. In contrast, cellular data tends to have 100-300 meter spatial precision. That means the Metrics we develop with Location-Based Services data are more fine-tuned to both visitors’ and Sites’ exact locations. (See Figure 2 below.)
- Ping rate: Devices using Location-Based Services generally send “pings” when the device changes location. In contrast, cellular tower “pings” are irregular. That irregularity makes algorithmically processed cellular tower data significantly less accurate.
- Privacy protection: Device users must proactively opt-in and enable Location-Based Services before they can use them. They agree that smart phone app providers can collect their locations. In contrast, cellular providers do not require proactive opt-in – and opting out requires consumers to contact their cellular providers.
Note that our Premium Site Metrics are still derived from navigation-GPS data.
Figure 2: Spatial precision matters when it comes to travel pattern analytics. As shown above, if device ping’s spatial precision is 300m, it is impossible to be certain of its location: it could be located in several different buildings, many different roads, and even in the middle of several different intersections.
What This Means for Our Customers
So, what does this really mean for our customers? A whole lot:
First, StreetLight InSight users can choose between navigation-GPS and Location-Based Services data for our most popular Travel Metrics, such as Origin-Destination and Zone Activity Analyses.
Second, our most popular Site Metrics – Home and Work Footprints, which describe the likely home and work locations of visitors to specific locations - are more accurate.
But we know that this choice could be tricky, so we’re here to help our customers select the best data source to use for the question at hand. While some analyses could be done with either source, certain analyses are better suited to one type of data. I’ll be writing up more guides on this topic soon. (Spoiler Alert: Mixing-and-matching or even blending Metrics derived from different data sources is typically the ideal approach.)
Generally speaking, however, this is what navigation-GPS and Location-Based Services data do best:
- Navigation-GPS data is best for analyzing trips’ characteristics, or how people get from point A” to “point B – or, as we say at StreetLight Data, Zone A to Zone B. Importantly, you can also differentiate between personal trips and commercial trips with navigation-GPS data. This data source does a great job of describing routes taken, including the relative volume of trips by route at different times and types of day, as well as the speed, duration, circuity, and length of trips.
- Location-Based Services data is best for analyzing groups of people doing an activity – like being at work or going out to lunch. It can also describe visitors’ likely home and work locations (in aggregate), and the volume of people that travel from Zone A to Zone B at different times and types of day.
Often, mixing and matching sources for a blend of analytics will give you the best possible results. We’re working to better automate mixing and matching as we learn from our clients how they want it implemented.
So, are you as excited as I am about our new Location-Based Services data? Click here to let us know - we would love to hear your feedback!