KEY CONCEPTS
• Vehicle telematics use telecommunications, vehicular technologies, electrical engineering and computer science to send, receive and store information through telecommunications devices to control, better operate and maintain fleet assets.
• Telematics can significantly improve fleet maintenance, but the primary challenge is not data collection but turning that data into actionable insights.
• A lack of clarity on the benefits and value telematics can deliver limits adoption and makes return on investment difficult to define. Many of the benefits of telematics come from avoided machine failures, which can be difficult to quantify in dollars. 
• Focusing on a small number of high-value signals, such as critical fault codes, is more effective than attempting to manage all available data. Identifying critical fault codes can allow failures to be caught early in the potential failure (P-F) curve and reduce repair costs.
• Pilot programs and strong user engagement are critical to identifying the right tools, building adoption and scaling successfully.

While the Internet of Things (IoT) has made connected technology widely accessible, many organizations and fleet managers are still working to translate telematics data into meaningful operational improvements. Telematics systems—combining onboard sensors, GPS and communications networks—can provide real-time visibility into vehicle health and performance. However, the shift from reactive to predictive maintenance depends less on the technology itself and more on how effectively organizations use the data it generates.

Vehicle telematics as a field includes telecommunications, vehicular technologies (e.g., road transport, road safety), electrical engineering and computer science. GPS technologies are integrated with computers and mobile communications technology. This allows information to be sent, received and stored through telecommunication devices that can control and better maintain remote objects like vehicles. 

In practice, implementing telematics is as much an organizational challenge as it is a technical one. Integrating data with existing systems, ensuring data quality and driving adoption across maintenance and operations teams are often more difficult than deploying the technology. Without a clear purpose and defined use cases, large volumes of data can quickly become overwhelming rather than useful.

This article draws on real-world experience from fleet professionals to examine how organizations are approaching these challenges. It focuses on practical lessons for starting telematics programs, engaging users and turning raw data into actionable insights that improve reliability, efficiency and decision-making.

Challenges to implementing telematics
STLE member Emlyn Eager, supervisor of fleet maintenance engineering and data for BC Transit in Victoria, British Columbia (BC), Canada, highlights that the primary technical challenges to implementing telematics are not just about collecting data, but about managing the data effectively. Large data volumes, system integration and sensor reliability all present significant hurdles. Data overload can be a major issue as decisions must be made between collecting and managing large data sets, versus focusing on metrics that could drive immediate action. In practice, organizations must strike a balance between capturing comprehensive datasets and focusing on the specific metrics that enable timely, actionable decisions.

A key complexity lies in integrating telematics data with existing enterprise systems. Computerized maintenance management systems (CMMS) are designed to plan and track maintenance activities. Aligning telematics inputs with these systems is often difficult, requiring either custom-built integrations or acceptance of limitations within existing software.

Eager also emphasizes the importance of distinguishing between actionable and passive data. Not all data generated by telematics systems warrants intervention, and organizations must decide whether to rely on automated, artificial intelligence (AI)-driven alerts or incorporate human validation into the decision-making process. This challenge is compounded by sensor reliability, particularly in harsh operating environments, where redundant data sources may be required to ensure accuracy. At the same time, enabling remote access to data introduces additional considerations around data governance, privacy and cybersecurity risk.

Jory Maccan, a reliability fleet analyst at Imperial’s Kearl Oilsands in Alberta, Canada, identified the biggest challenges facing organizations implementing telematics to be “the workforce understanding what they are, how they work and how much power there is in having access to real-time data and information on operational parameters.” He finds “that old school maintenance people stick to what they know works—sight, smell and feel when troubleshooting or repairing.” 

Maccan found that understanding the analytics available and learning what they can and can’t do were “some of the hardest parts of the whole operation.” Issues like “communication problems and software incompatibility are easy to deal with when you have the right technology people in place.” In contrast, Maccan finds that “getting the people who have been working in the field to recognize the new avenues open to them in monitoring and repair has been the harder challenge.” 

Ahmed Mumeni, a program manager transit zero emission lead at Jacobs in Victoria, BC, Canada, agrees and identifies the biggest challenge to telematics as the poor understanding people have of what it offers. Initially, Mumeni finds organizations “view telematics as GPS tracking of vehicles or assets, without a clear connection to maintenance reliability or operational efficiency.” To overcome this challenge means “connecting telematics with clear outcomes (e.g., reducing idle time, improving the ability of technicians to improve diagnostics or gaining visibility over how equipment is being used).”



A trial telematics program
Hayden Chuter works as an electrical maintenance engineer with Coast Mountain Bus Company (CMBC), working out of the Burnaby Transit Center in Burnaby, BC, Canada. TransLink/CMBC completed a trial telematics program in 2024 and 2025. Beginning in 2026, the center is beginning an implementation project using the lessons learned through the trial program. Chuter notes, “At this time, we do not have an operational telematics tool; however, we do have a very defined structure we’re going forward with.” 

Chuter identified three major challenges faced when first implementing a telematics or connected device platform:

• The large quantity of vehicles to be connected. The CMBC operates bus transit services in metro Vancouver, BC. TransLink is the metro Vancouver’s transportation network. At the time of writing, CMBC/TransLink operates a fleet of over 1,700 vehicles. Chuter notes that “deployment of solutions to a large fleet is challenging and expensive, especially with the multitude of projects looking to deploy new technologies.” To mitigate this issue, Chuter says they are running the “system as a software package onboard our new vehicle gateway, so net zero new hardware is required.” 
• Date privacy/vendor secrets. While the CMBC/TransLink vehicles contain sensors and significant data, Chuter says “the decoding of the networks is a closely guarded secret as the vendors offer their own telematics systems.” To navigate this issue, Chuter says they have “been utilizing the open J1939 database working with vendors over time to increase availability of ‘proprietary’ information.” The J1939 CAN standard is a communication protocol designed for heavy-duty vehicles and machinery.
• Individualized user needs. Chuter notes the different user groups all need different data displayed in different ways. These groups include maintenance, engineering, supervisors and dispatch, among others. The CMBC/TransLink system is using Microsoft Azure Cloud and Power BI to create a flexible platform that can quickly generate new dashboard displays to meet user group needs. 

Chuter says that once data are collected, translating the raw data into actionable insights for fleet maintenance and performance optimization requires taking advantage of subject matter experts (i.e., fleet maintenance engineers) “to help understand available data, and how it can best be captured and utilized.” The engineers “have special knowledge in vehicle networks and sensors and full understanding of available data.” This allows them to work “with our business intelligence analysts to transform the data into usable formats, better suited to different stakeholder needs.” 

To ensure that different teams (e.g., maintenance, operations, management, etc.) use the collected data effectively to improve decision-making, Chuter’s “project team has a structure established to regularly meet with representative stakeholders, and to iterate on dashboards and tools utilizing the Agile methodology to refine tools based on stakeholder/user input.”

Coordination challenges
Devin Fitzsimmons, fleet maintenance engineer with BC Transit in Victoria, BC, Canada, notes his organization’s biggest challenge “has been signal mapping and lack of plug-and-play solutions that we were hoping for.” BC Transit has a mixed fleet from different bus vendors that all use the J1939 CAN standard. However, despite using the same standard, BC Transit’s “telematics vendor has not been able to map the signals we originally included in our scope of work.” As a result, the scope has been scaled back to very basic information (e.g., odometer, engine hours). Another challenge has been a lack of control over the data or visuals, where changes can take months to implement. 

While a dispatcher may need to know real-time location and vehicle status, a maintenance technician may only be interested in fault codes to support diagnostics. Mumeni notes that “all teams need to work with the same data source, and they should be able to make operational decisions using the same platform, but with role-specific capabilities.” For these teams to use the same telematics platform, the available data must be easier to access and use than the status quo of not having visibility over equipment status. Mumeni says, “There also needs to be a concerted change management effort in place to support implementation and sustainment of a telematics system.” 

Fitzsimmons notes, “Bus vendors should be more open with available data onboard the bus and realize that pay-walling critical signals behind their own telematics suite does not serve the customer when most agencies have mixed fleets.” 



Costs and cost benefits
Mumeni notes, “There are lifecycle costs to consider when implementing a telematics system.” He says, “Hardware, installation, data plans and subscription fees need to be justified in a business case, and a return on investment is challenging to quantify.” He agrees that small scale pilot tests can be used to validate the value of a telematics system before pursuing a large-scale deployment.

However, Mumeni finds the lack of an “industry benchmark for what a successful implementation of telematics or its benefits could look like” results in the benefits of applying telematics not being immediately clear. He finds “the value of telematics is through avoided machine failures, or through the ability of frontline operations to make better decisions; this can be difficult to quantify in a business case.” Defining key performance indicators (KPI) early is key to overcoming this challenge. Examples of KPIs can be idle time reduction and identifying critical fault codes ahead of catastrophic failures.

Best practices when starting out
Mumeni says, “Organizations seeking the implementation of a telematics system need to start with a clear purpose.” Defining the operational or maintenance problems the organization is trying to solve before deploying the technology is critical. He finds that “telematics delivers the most value when it is aligned with concrete goals (e.g., reducing idle time, improving reliability engineering efforts by making more granular data available).” He agrees with Chuter that “small pilots can validate assumptions and demonstrate value without a large upfront investment.” Mumeni also recommends including frontline operations and maintenance staff in the development phase, so they understand the specific requirements and benefits a telematics system offers for easier adoption. 

In addition to having a clear purpose, Chuter also recommends organizations just beginning their telematics/connected device journey “start small and try to do a proof of concept first.” This approach helps the organization understand what is needed, how it should be displayed and how it can add business value. 

Also, Mumeni says, “A successful telematics program needs a deliberate change management effort.” Having clear executive sponsors who articulate why the organization is investing in telematics is critical. In addition, having “change champions within maintenance and operations teams helps to reinforce a data-driven culture and supports a wider adoption and sustainment over time.” Chuter agrees change management is very important. 

Mumeni says, “Data governance is an emerging area that organizations are grappling with.” Questions about data definitions, ownership, access and long-term stewardship need to be addressed as the volume of data increases and telematics adoption grows. He says, “Clear governance frameworks are required to identify data owners, and seemingly simple ideas like defining what a data point means, or how it’s used in an enterprise” become important.

Transforming raw data into actionable insights
Telematics systems can generate a lot of data quickly. Maccan notes that the tools most helpful in transforming raw data into actionable insights aren’t tools, so much as people. People must use the tools regularly. Telematics designers need to “push people to at least try using the tools regularly, ask them how it worked, ask them if they have ideas to make it better, then go back to the run/design team with those ideas to tailor the experience and report to the people that will use it.” Maccan recognizes that while a developer can have lots of ideas about both data and tools, “at the end of the day, the end-user will know best what works, how to present it and what they need.” 

Ensuring that different teams (e.g., maintenance, operations, management, etc.) effectively use the collected data to improve decision making is also critical to success. Maccan finds the key point is ensuring they “are aware that the data exists or they won’t go looking for it.” In fact, the teams need “a tool that fits their needs in order to want to use it, and they have to feel like their asks are being reflected in changes to the tool, [and] new reports that are being developed.” In addition, Maccan says, “They have to be allowed some time to learn it, learn what they can do with it and provide requests and feedback on it.” Unless the tools fit the end-users needs, “it’s just an inconvenience in a busy day.” He notes that tailoring the tool to the user’s needs is more effective than mandating use of a tool that doesn’t fit the user.

Fitzsimmons states that strategies and tools to help transform raw data into actionable insights for maintenance and performance optimization are still a work in progress. The telematics portal’s user interface has limited the ability to gain fleet-wide insights. He notes that his organization is still putting “time and effort into just validating the integrity of the data presented in the telematics portal” and are not yet to the point of effectively using these data to improve decision-making. An example of the complexity they face on the data end is that each bus can have multiple odometer signals that don’t match, while different types of buses can have different sources for those signals. This complicates making critical decisions using these data as decisions must be made regarding which signal to follow. 

Mumeni notes, “Organizations implementing telematics devices need to understand the problem they are trying to solve before collecting or analyzing data.” Without a clearly defined operational goal, the large volume of data telematics platforms can generate can confuse translating the data into meaningful action. A potential starting point could be identifying a specific question to answer. For example, where are vehicles or equipment idling the most? By clarifying organizational objectives, the specific raw data needed “can be collected, transformed, structured and visualized in a manner that would support operational decision-making.”

Mumeni finds that the “use-case specific dashboards (as opposed to generic reporting) can be helpful to drive specific operational outcomes.” For example, if an organizational goal is to reduce vehicle idling, a dashboard that aggregates data to highlight idling behavior can be created.

Timing of adoption
Maccan notes a major timing issue is related to the technology where “connected devices, data integration and AI are rapidly moving toward the category of foundational tools rather than optional add-ons.” Whether an organization is an early adopter or waits to learn from others may shape that organization’s position in their field going forward. While moving first has risks, Maccan says, “It also provides the opportunity to actively shape systems, workflows and standards to fit specific operational needs, as well as be the first across the line with a finished product that adds value.” Additionally, he finds that “early adopters gain institutional learning, have influence on best practices and rapidly develop internal expertise that is difficult to replicate later (living through it is a better teacher than watching/judging someone else going through it).”

In addition, Maccan says, “Organizations that delay adoption may save on early missteps but risk inheriting solutions designed around someone else’s constraints rather than their own, and still have to put development time into tailoring it to their operation, while the others are taking the next big step.” While waiting may reduce risk, it also can mean a loss of both flexibility and influence on the resulting outcome. He finds it “important not just to adopt new technology, but to at least consider doing so deliberately—pairing experimentation with governance, flexibility with accountability and innovation with problem-solving capacity.” Successful organizations are those “willing and able to engage early, adapt quickly and continuously refine how technology is used to deliver real value.”

Fitzsimmons recommends that organizations just beginning the telematics journey select a telematics vendor that is well established in the particular field. He recommends running trials and being “clear on the goals and requirements from internal stakeholders.” Listing those internal requirements in technical specifications prior to looking for telematics solutions and for new vehicles is critical to success.

Conclusion
In terms of best practices, Maccan feels that the most important best practice to recommend to organizations just beginning in telematics is to “have the frontline people tell you what they need—involve at least some of them in the selection and development.” Giving frontline people the opportunity to be involved from the ground level of development means creating champions and power users for the next step in the digital evolution of the project. Maccan finds this step is often missed with tools researched and implemented by people who will never actually use them.

Also, Maccan notes that getting people to use telematics tools can be the biggest challenge in implementing these programs. He suggests: “Let people try it, complain about it, then solicit their ideas, and implement those ideas—users will make better use of a tool that is exactly the right tool, than a tool that’s close enough but inconvenient.” He says that the “voice of the customer is key.” 

Note that telematics can be used in areas beyond fleet maintenance, including inside factories that use industrial equipment (e.g., forklifts) where telematics can still track equipment locations. Eager notes that telematics can also be used for stationary equipment where key data points can be transmitted and used for predictive maintenance. 

Additionally, while telematics can significantly improve fleet maintenance, the primary challenge is not data collection but turning that data into actionable insights. Identifying critical fault codes can allow failures to be caught early in the potential failure (P-F) curve and reduce repair costs.

While potentially challenging to implement, telematics programs offer the possibility to improve efficiency, while saving time and money for organizations that implement these programs effectively. Many of the benefits of telematics come from avoided machine failures, which can be difficult to quantify in dollars.

ADDITIONAL RESOURCES
1. www.nationalacademies.org/read/29236/chapter/1
2. https://theusajournals.com/index.php/ajast/article/view/7123/6585

Andrea R. Aikin is a freelance science writer and editor based in the Denver area. You can contact her at pivoaiki@sprynet.com.