In our recent webinar, Reliability Beyond the Numbers: What Monta and Kempower’s Integrated Tech Stack Unlocks for CPO Einride, we explored how smart and supported hardware-software integration is shaping the future of EV charging. By bringing together DC fast-charging leader Kempower and trusted CPO partner Einride, we unpacked how collaboration, high quality integration, and ongoing improvements and updates are transforming reliability, performance, and efficiency across charging networks.
Watch the session on-demand here.
Monta: Powering millions of charging sessions
Monta is the fastest growing software platform in eMobility with over 240,000 connected chargers, 2.5M sessions per month, and 1M registered users (320k active monthly). This provides a unique dataset and perspective on the technology stack CPOs across Europe are building on. As an end-to-end platform providing the full technology stack, Monta is able to see, monitor and address issues across a broad range of failures. This dataset, alongside observing differences between over 700 different hardware models integrated into the platform, provides the foundation for our discussion with industry experts Emma and Rolle.
Monta sees a 3x difference in performance between the best and worst performing hardware integrations. From circa 5% to 15% failure rates respectively
Introducing our guests
Rolle Nieminen – Kempower
Rolle, Kempower’s Head of Sales and Country Manager, brings deep experience in EV infrastructure and the automotive sector. Kempower manufactures DC fast chargers and provides software and services on top of its hardware, with deployments in over 60 countries.
Rolle highlighted that Kempower is constantly working on and updating their OCPP stack – it’s certainly not a one-and-done process. Close collaboration between hardware manufacturers and software providers is crucial for maintaining reliability and delivering a seamless user experience.
Emma Lindberg – Einride
Emma has been with Einride for over three years, focusing on charging infrastructure as part of the company’s digital freight platform. Einride operates its fleet and provides charging as a service, managing both private and public charging stations.
Emma stressed how critical a high-quality OCPP integration is, not just to avoid failures but to unlock innovation. For example, because Einride didn’t need to spend time troubleshooting OCPP fixes between Monta and Kempower, they were able to build an energy-based reservation system, propelling their fleet service forward.
The role of OCPP in real-world charging
OCPP 1.6, the most popular protocol that aims to standardise communication between chargers and backend systems like Monta, has been around for 10 years. While it provides a standard messaging framework, it’s not a tightly prescriptive specification, meaning implementation can vary between chargers and CPMS platforms. This variation can often be the cause of failed charge sessions, and error prone infrastructure with inconsistencies meaning, transaction timeouts, inconsistent error handling, and variations in expected offline behaviour. All of which can result in a frustrating driver experience, and hurting infrastructure ROI.
At Monta, we value quality and see the importance of running extensive testing on hardware, new firmware, timeout handling, and more, not just a simple protocol check. Monta has a dedicated Integrations team purely for handling these kinds of critical touchpoints between CPMS and hardware. This ensures reliability and high performance across complex charging networks. Monta also provides specific tooling for Hardware Manufacturers with assets in the platform to give them AI powered troubleshooting and deep data insights.
Monta’s platform average of charge success (excluding Roaming and 0kWh sessions as failed) is 95%. Einride is a high performing network with over a 97% charge success rate (even when counting 0kW sessions as failed) across Einride’s fleet. This is especially critical for Einride’s business model as not only are they making significant investments in charging infrastructure, they are also dependent on it themselves as a Freight-as-a-Service provider. Failed or slow charges means increased fleet downtime, and heavy impact on customer SLAs. Ensuring thorough integration and consistency to drive reliability has been a key focus enabling Einride to scale and innovate rather than fixing simple OCPP issues.
Lifecycle of a charging session: example of where failures can occur
- Vehicle handshakes: vehicle times out or throws an error in this step
- Payment authorisation: RemoteStart rejected for some reason, invalid token
- Charger transitioning to active state: no StartTransaction within connection interval
- Communication timeouts: missing or delayed StartTransaction
- Malformed Data: Bad Time or Meter Stamp, or StopTransaction
Emma explained how Einride addresses these challenges:
- Ensuring chargers meet high technical standards through testing prior to launch, and rigorous re-testing.
- Using Monta’s ‘alerts’ system and scheduled restarts to reduce failures.
- Building processes that allow operations to focus on higher-level services instead of troubleshooting basic failures.
- Standardising on Auto-Charge through Monta to streamline authorisation
This structured approach allows drivers to just plug in and charge, a core requirement for reliable operations.
Emma and Rolle both emphasised that innovation and daily operational excellence must go hand in hand. Reliable OCPP integration frees up resources for building advanced services – like energy-based reservation systems – while AI can further assist in network monitoring and predictive maintenance.
This area – OCPP integration – is often an afterthought, but it’s critical to high-performing infrastructure. Focusing on it enables not only reliable operations but also innovation, like the tools Einride has built on top of a stable foundation.
The application of AI
AI is increasingly becoming a critical tool in managing EV charging networks, supporting both operational efficiency and strategic decision-making. Discussion with our guests, and feedback from our audience highlighted several ways AI can add value:
- Active but controlled assistance: AI can help operators manage day-to-day operations without taking full control, allowing teams to focus on higher-value tasks.
- Revenue and customer satisfaction: By enabling operators to quickly understand and resolve failed charges, AI helps improve the overall performance of the network, ultimately increasing revenue and driver satisfaction.
- Automation and workflow support: AI automates routine workflows and provides step-by-step guidance for complex processes, freeing up time and relieving stress for operator teams.
- Reduces need for specialist expertise: AI removes the necessity for deep knowledge in OCPP or data science, allowing teams to assess and understand network performance criteria without relying on specialist skills.
- Support for complex standards: AI can assist with code generation and interpretation of standards like OCPP 2.0.1, accelerating implementation while maintaining reliability under human oversight.
Audience poll results
- What is the path of least resistance to improving reliability in the medium-term? : 62%: closer collaboration with stakeholders, 17%: predictive maintenance, 14%: ISO15118 and better vehicle comms, 5%: more resources in Network Operations, and 2%: using OCPP 2.0.1.
- How active do you think AI could/should be in Network Operations? 71%: somewhat – I’d like AI to be active but limited, 24%: very – I want AI to have the power to do everything a human can in a network, 5%: passive – I don’t want AI to be able to complete any actions, just tell me what I need to know, and 0%: not at all.
- Do you think AI will help make OCPP1.6 more usable, and slow down the adoption of 2.0.1, or will the increased data and accuracy be such a gift to AI agents that it speeds up the shift to more advanced standards? 73%: no – 2.0.1 is coming in 2026, AI will love it, 13%: yes – AI will slow the adoption of 2.0.1 by improving 1.6.
In short, AI is a powerful enabler, not a replacement. When combined with solid integration, strong collaboration, and robust operational processes, AI can enhance reliability, streamline operations, and unlock new opportunities for innovation across charging networks.
Key takeaways
- Reliable OCPP integration is foundational: It ensures smooth operations, reduces failures, and enables innovation.
- Collaboration drives performance: Hardware and software teams must work closely to deliver consistent user experiences.
- Data and insights matter: Monta’s platform provides the transparency and metrics needed to optimise charging operations.
- AI is a tool, not a silver bullet: It can accelerate processes, but human expertise and rigorous testing remain critical.
- Medium-term improvements are achievable: Focusing on collaboration and operational best practices yields tangible results faster than chasing new protocol versions alone.
- Reliability doesn’t happen by accident: It’s built through the right integrations, data, and operational tooling.