Transit agencies aim to efficiently deliver service and improve the rider experience. High-quality data is critical in meeting these goals, but many agencies simply can’t evaluate their own data quality or implement solutions to improve it.

We recently hosted a conversation with two transit agency leaders on the front lines of solving some of the most pressing challenges for data quality. In this webinar with Valley Metro’s Michael Pal and Kansas City Area Transportation Authority’s Tyler Means, along with Swiftly data quality expert Lauren Grabowski, we covered common data quality issues and how they impact service reliability and the rider experience.

You can watch a recording of the session here, or read on for a framework for thinking about data quality and highlights from the conversation.

The three dimensions of transit data quality

Data quality has many different factors that impact the success of staff- and passenger-facing tools in multiple ways. With so many interconnected parts, it can be difficult to know where to start with improving data quality. Let's start with a framework for evaluating the different components of data quality.

We think of data quality across three categories:

1. ‍Data accuracy: The degree to which transit data correctly reflects actual service, ensuring that information—such as vehicle locations, arrival/departure times, on-time performance (OTP), and predictions—matches real-world performance.‍
2. Data completeness: The extent to which all relevant transit data is captured, including both delivered and undelivered service, and whether service adjustments and disruptions are properly recorded and communicated to riders.‍
3. Data accessibility: The ease with which transit data can be retrieved, understood, and used by stakeholders, ensuring data is available for analysis, decision-making, and communication.

At Swiftly, data is at the core of everything we do, which is why we focus intently on measuring and enhancing data quality across these three dimensions. Next, let's walk through the most common strategies transit agencies can take to improve data quality in each of these categories.

1. Data accuracy

Better location tracking

Accurate, reliable location tracking is key to high-quality transit data, yet most legacy systems rely on a single AVL source, updating only every 30+ seconds. Swiftly integrates multiple AVL feeds—including vehicle routers—delivering updates as fast as every 5 seconds for more precise tracking and richer historical data.

Redundant AVL sources also improve system resilience, preventing data loss from operator errors, equipment failures, or CAD/AVL outages, ensuring continuous real-time and historical visibility. As Tyler Means, Chief Mobility and Strategy Officer at KCATA, said:

“We went through a cyber attack in 2024, and it lasted for about 5 weeks. We lost access to our CAD/AVL system, so our dispatch and departments who track our system throughout the day were not able to see where things were.

“We learned that onboard routers were still providing location data, and Swiftly had integrated the inputs and was utilizing that as the real time information. We took some clean computers and ran our dispatch off Swiftly, so that they could actually follow our fleet. We were able to actually get back up to speed after only a few hours and monitor our system utilizing Swiftly.”

Improving predictions

Agencies have typically struggled with understanding and improving the accuracy of real-time predictions provided to riders. To address this, Swiftly partnered with Transit to develop the ETA Accuracy Benchmark, an industry-standard methodology to measure the precision of real-time Estimated Time of Arrival (ETA) predictions provided to riders. This allows agencies to evaluate their ETA accuracy, track improvements over time, and benchmark against peers.

“What we really focus on when we look at our on street experience is, ‘How do we ensure that they know a bus is coming and it's coming on time?’ That was really kind of the emphasis of us wanting to improve our prediction.

“We successfully improved our prediction accuracy by 22% and reduced miss-the-bus predictions by 94%. As a result, we've cut down on those missed trip calls by over 10%. So we're seeing people being able to just access the bus more successfully.”

—Tyler Means, Chief Mobility and Strategy Officer, KCATA

Beyond geofences

Legacy systems for measuring OTP often rely on geofences to determine stop arrivals and departures. However, geofences can be error-prone, leading to inaccurate data. For example, if a vehicle departs a stop but doesn’t immediately cross the geofence, then the recorded departure time may be incorrect. These inaccuracies not only distort OTP and run-time data but also sow distrust in metrics. In contrast, Swiftly’s advanced algorithms, which analyze GPS data and historical patterns, produce more accurate arrival and departure estimates, resulting in more accurate OTP reporting, better schedules, and metrics that all stakeholders can rely on.

Mike Pal, the Chief Transportation Officer at Valley Metro, recently shared a post describing how Valley Metro and Keolis, their private contract operator, used Swiftly to gain a common understanding of current OTP and to improve this critical metric from mid-70% to ~85%. He shared, “We had discrepancies in data from geofencing and other issues. There were challenges and endless meetings and disputes about assessing liquidated damages with our contractor based on what the true on-time performance was. We went out into the field with our legacy system and Swiftly at the same time and measured.”

“We saw Swiftly was accurate, and the legacy system could not calculate the OTP because of the geofence confusion. Once we started relying on Swiftly for our OTP measurement, we really built trust between our contractors and Valley Metro.”

2. Data completeness

Filling assignment gaps

Missing vehicle assignment data—often caused by operator error, hardware issues, or data feed problems—can create gaps in real-time passenger information and historical performance data. Swiftly's Auto-Assigner automatically matches unassigned vehicles to scheduled trips, ensuring real-time locations and actual arrival and departure times are accurately recorded, even when assignment data is incomplete. This enables agencies to collect 10–20% more historical data, enhancing passenger information and OTP reporting.

The Data Completeness Report

OTP is a key metric for transit agencies but fails to capture critical factors like ghost buses, uncommunicated service adjustments, and data gaps. Swiftly’s Data Completeness Report complements and expands on OTP by integrating service delivery and communication insights, helping agencies identify gaps, improve reliability, and enhance the rider experience.

Tyler Means of KCATA shared, “We really focused on eliminating ghost buses in a couple scenarios. First, did the bus actually leave the barn to get out in the field? Or was there a detour? The bus is out in the field, but it’s not going to your stop. If you’re not serving that trip or transmitting that information to riders, you're almost lying to them. You're going to work twice as hard to get that rider back.”

"So what happens now is that dispatch can put a detour into Vontas, and that feeds directly into Swiftly and updates automatically to the rider. So a rider will know that the bus is no longer coming to them. They can see that their stop is getting passed by, and they can see the stop that's closest to them to access the ride. Through that integration with Swiftly and Vontas, we're able to make sure that riders have the same information we do.”

3. Data accessibility

Intuitive web-based dashboards

Accessing performance data from legacy systems is often a challenge, with agencies limited to pre-set reports or needing complex access methods like VPNs. In contrast, agencies using Swiftly can easily access performance insights through a well-designed, intuitive web-based dashboard. With unlimited user logins, all departments—including operations, service planning and scheduling, customer service, and executives—can access the data, promoting transparency and better decision-making across the organization.

Data access in real time

Most legacy systems have a delay of three days or more in processing performance data, making it difficult for agencies to respond to service issues in a timely manner. In contrast, Swiftly provides real-time access to performance data, empowering agencies to monitor and adjust service as it happens. This proactive approach is essential for improving service reliability and the rider experience. As Valley Metro's Mike Pal said:

“It’s nice to work in the present rather than 3–7 days in the past. You can’t fix performance once the day is over. You have to actively manage and be proactive, not find out about problems days later. Swiftly helps us do that.”

Unlocking functionality with APIs

Swiftly further enhances accessibility by offering APIs that empower agencies and their partners to unlock additional functionality:

• ‍Mobile apps: Platforms like Google Maps, Apple Maps, and Transit can deliver accurate real-time vehicle locations, ETAs, and service alerts to riders. Transit also leverages Swiftly's GTFS Trip-Modifications API to display detour shapes.‍
• Agency staff: Agencies can integrate Swiftly data with other sources to create custom dashboards, such as combining operational and financial data in executive dashboards.‍
• Consultants: Swiftly data supports transit planning projects, including network redesigns and corridor studies.‍
• Vendors: Integrations powered by Swiftly data enhance agency operations, such as optimizing schedules using running time data and identifying and implementing Transit Signal Priority (TSP).

Swiftly is on track to process over 7.4 billion API calls per year, enabling powerful integrations with tools that benefit agencies and their riders.

“We’re a big city, and we run buses over a massive area. We can't put real-time signage everywhere. By pushing real-time passenger information into a mobile app, such as Transit or any other third-party app, we're able to put real-time signage in people's pockets.”

—Tyler Means, Chief Mobility and Strategy Officer, KCATA

Empowering transit agencies with high-quality data

Data quality can feel like an abstract issue, but, as we've heard from transit agency experts, has very tangible impacts on service reliability and the passenger experience. With high-quality data, transit agencies can focus their energy on projects and activities that will drive the greatest impact for their communities. Agencies that have invested in data quality are positioned to make inroads in the following areas:

• ‍Enhance service reliability: Accurately monitor vehicle locations, OTP, and service disruptions to reduce missed trips and improve schedule adherence‍
• Improve rider communication: Provide reliable real-time passenger information, such as ETAs, service alerts, and detour updates, enhancing the rider experience.‍
• Optimize operations: Identify inefficiencies, adjust schedules, and allocate resources effectively to reduce delays and improve service management.‍
• Track performance: Measure service metrics like OTP and service delivery more accurately, enabling agencies to monitor trends and make data-driven adjustments.‍
• Support strategic planning: Use data insights for route planning, service expansions, and infrastructure investments based on actual performance trends.‍
• Increase transparency and accountability: Share clear, accurate data internally and with the public, promoting trust and collaboration among stakeholders.‍
• Enable technology integration: Leverage APIs and data platforms to power third-party apps and tools, and analytics dashboards for enhanced functionality.

In short: High-quality data helps transit agencies deliver better service, enhance the rider experience, and make smarter operational decisions.

Hear more from agency executives

If this topic resonates, make sure to watch the recording of our webinar on data quality to hear firsthand how two transit leaders are tackling this challenge.