Published counts or mass mobile data? — how to select the right traffic data
In this article we investigate the implications of the data you select and how to make the right decision for your business.
Big data vs small data – how much is enough?
Scientific data modeling is an accepted practice in the traffic data industry. It’s impossible to track every single car on every single road — so data from more manageable sample areas are collected and machine learning and AI are used to take the original data sample and apply the outcomes to all roads, resulting in a full data set.
Both mass mobile data and published counts require some amount of data modeling to ensure that they represent the full population, but they’re collected in very different ways.
Mass mobile data is gathered from cell phone signals on the road. Not every vehicle can be tracked with mobile signals, so assumptions are made on the actual number of vehicles present on the road. Published traffic counts are collected from local transportation departments, who use devices to track and count every car on a given road, but this isn’t available on all roads.
While data modeling is used in both cases, published counts begin with actual traffic count data, giving the data modeling tools an accurate traffic count to inform machine learning algorithms.
As privacy laws evolve and the public become more aware of their privacy rights, and potential violations, more and more cell phone users are restricting tracking on their devices. This causes a problem for mass mobile data as the sample sizes are steadily reducing. In contrast, the consistency of published traffic counts is protected as there is a federal regulation in the US that requires each state to submit published traffic counts as part of the annual Highway Performance Monitoring System (HPMS) report.
All traffic data uses some form of data modeling. It’s important to understand the original data sources and sample sizes to ensure you have representative data that you can base decisions on.
What matters most — vehicles or the people in them?
Different industries require different outputs from their traffic data which can inform the best type of data to choose.
For example; industries that rely on trade from passing vehicles, like gas stations, require data that shows them the number of cars that pass on the streets around their location. Other industries like restaurants, require data on the number of people that pass the streets around their location.
Mobile data will provide information on the number of people in a passing vehicle — not the number of vehicles. It’s likely to include people in buses and on bicycles, so can provide great insights to a restaurant business who wants to understand their potential passing trade.
For gas stations, AADT and published counts are more reliable. Published counts relate directly to the number of vehicles on the road, regardless of the number of people within the vehicle.
Many businesses may require a combination of both types of traffic data. Restaurants and retailers would usually require mobile data to understand the number of people in their location — but the increasing popularity of drive throughs is changing this.
Greg Rutan from Trade Area Systems suggested “Drive–throughs have always been prevalent, but their popularity will increase in a post–covid world – we’re used to drive through restaurants, but we now have pharmacies, and even cannabis shops. In the future will see other industries adopting drive–throughs as people don’t want to leave the safety of their cars to shop in crowded areas anymore.”
For retailers and restaurants that operate or plan to open drive–throughs, combining published traffic counts with mobile data will give the most reliable representation of both people and vehicles in your location.
Understanding your objectives will allow you to effectively chose the right data for you. For businesses that require multiple data sets, tools like TAS can help to quickly combine and visualize data for easier analysis and decision making.
Understanding the limitations of traffic data.
Published AADT counts are widely regarded as the most accurate traffic data type available. However, the nature of the collection methods mean collection is time consuming. The US Department of Transportation require published counts from every state annually, which mean that published counts reflect figures collected in the previous year and so may not account for any recent changes in traffic behavior.
Mass mobile data tends to have smaller original sample sizes, so it relies heavily on data science and machine learning to ensure that it’s representative.
This blog post, Traffic data that provides genuine intelligence: why quality is key, will help you to examine critical aspects of traffic data to ensure you select traffic data with integrity.
Make the right decision.
Every business will have different needs and it’s important to understand your requirements and objectives before deciding which type of data is right for you.
Machine learning and AI is only as good as the data that feeds it. In order to base strategic business decisions on any data you need to ensure it has come from a credible source.
TrafficMetrix® provides users with verified published traffic counts. Collected from trusted sources, including the DoT, it’s combined and consolidated into a user-friendly format, ready to analyze as raw data or upload into any GIS tool.
With full coverage across North America, it’s the traffic data provider of choice for anyone looking for reliable AADT or published traffic counts.
Found this interesting? Why not share it:
Read more articles about:Data intelligence
Subscribe and get the latest updates
You may unsubscribe from our mailing list at any time. To understand how and why we process your data, please see our Privacy & Cookies Policy
Are you getting the most from your Fuel Site Analysis?
We recently launched a new range of Gas station Feasibility studies which give you even more actionable insights....
Fuel Pricing self assessment checklist
This self assessment will help you to gauge current capabilities within your Fuel Pricing team.