From Fragmented Data to Proactive Transit Operations
TAM America had the privilege of conducting an interview with John Maglio, CEO, ETA, and is pleased to share his insights on why transit agencies need a connected data foundation, how AI fits into that vision, and what more proactive transit operations could look like.
How has your thinking about what transit agencies need changed over the course of your career?
I've spent over two decades selling technology into transit, and for much of that time I was working within the same model as everyone else: every vendor builds a system that answers its own slice of the operation, and no one is responsible for the whole. Those tools solved real problems, but each one added another island of data no one else could reach. Agencies ended up with more dashboards than ever and still couldn't get a straight answer, because the data was split across systems from different vendors that were never built to talk to each other and staff were left to piece the story together themselves.
That changed how I think about what transit software should do. The goal shouldn’t be to give agencies another place to look. It should be to bring their information together so they can get to answers faster and make better decisions. That’s what we’re building with TransitOS, a Transit Operating System designed to connect the systems and data an agency already relies on and create a foundation for more proactive operations.
What does that look like in practice? How does TransitOS work?
TransitOS sits across the systems an agency already uses and brings that information together —scheduling, maintenance, workforce management, CAD/AVL, and other operational or enterprise systems. We connect through available APIs, existing standards, and read-only database access. The point isn't to replace what an agency has. It's to make the systems they already paid for finally act like one.
Think about something as basic as trying to understand why service deteriorated on a route. The answer might involve schedule adherence, operator availability or the condition of the vehicles assigned to that service, with each piece of information living somewhere different. Today, someone may have to go into several systems and piece that story together manually.
Once those systems are connected through TransitOS, that information can be evaluated together. TransitGPT is the first application we’ve built on top of that foundation. It allows staff to ask a question in plain English and get an answer using information from multiple parts of the operation.
Transit agencies already have a lot of systems in place. Why has it been so difficult to connect all of that data?
A big part of the challenge is that these systems were never designed to operate as one environment. Agencies often have platforms from different vendors, each with its own data structure and way of sharing information.
Standards like GTFS and SIRI have helped tremendously, but they only cover certain types of data. A lot of important operational information still sits in systems with limited APIs, proprietary formats or no standard interface at all.
So standards are part of the answer, but they aren’t the whole answer. If you want a more complete view of an agency, you also have to be able to work with the systems and data that fall outside of those standards. That has been one of the harder problems for the industry to solve.
AI is getting a lot of attention across transit right now. Where does it fit into what you’re building?
AI only becomes useful if it has access to the right information. That is why we think the data foundation has to come first. If an AI tool can only see one system or one slice of the operation, its answer is limited by what it can see, even if the answer sounds complete.
For us, the bigger opportunity is what becomes possible once AI can work across a much broader picture of the operation. Answering questions is the first step, but it is only the beginning. We frame the progression as Evaluate, Trigger, Act, or ETA. First, the system evaluates what is happening. Then it can trigger attention or a workflow when something needs action. Ultimately, it can begin taking appropriate action itself.
That is how we see AI changing transit software. Instead of being something people constantly have to monitor, it becomes something that actively helps them run the operation.
Agencies are under pressure to do more with limited staff and funding. How does that change what they should expect from their technology?
It raises the bar. When agencies are short on people and operating under tighter budgets, software cannot just give staff more information to monitor. It has to reduce the amount of manual work required to understand what is happening and decide what needs attention.
That is a big part of what is driving our thinking with TransitOS. If the system can evaluate information, surface issues automatically and increasingly help with the response, then technology starts giving staff time back instead of adding another tool they have to manage.
I think the question agencies should be asking is changing. It is less about how much a system can show you and more about how much work it can take off your team’s plate.
What do you think transit technology needs to look like five years from now?
Five years from now, agencies won't waste time poring through dozens of reports, interpreting dashboards, or stitching data together in Excel. They'll ask a question, and the system will assemble what they need in the moment. This can be an answer, an alert, a visualization, or even a dashboard built for that one question.
That's the direction we're building toward with TransitOS: software that can Evaluate, Trigger, and Act across the operation. The dashboard becomes an output that the system builds when it's useful, not the place people go to do the work.
Technology should be doing the work of finding what matters, helping determine what should happen next and, where appropriate, taking action. That is a very different future from simply giving people more information and asking them to keep watching the screen. If you would like to discover how ETA’s TransitOS software can improve your operations then please contact us. We look forward to hearing from you!










