Utilities are facing aging infrastructure, evolving regulatory landscapes, and the urgent need for decarbonization. As a result, many have discovered that their greatest challenges as well as opportunities lie not in field operations but in the back-office systems that power decision-making. From grid modernization to distributed energy resources, the transformation reshaping utilities requires more than just new technology. It requires a fundamental reimagining of how energy companies operate, plan, and scale.
Kunal Saxena, a senior management professional with extensive experience in the energy, power, and utilities industry, has guided organizations through some of the sector’s most complex financial and operational transformations. An advisor to C-Suite executives on technology innovation and industry trends, Kunal specializes in optimizing business processes to reduce operational costs while preparing companies for disruptive technologies like Generative AI. His work focuses on helping organizations navigate key industry challenges including decarbonization, utility decentralization, and creating scalable service delivery models.
Kunal shared insights with us on how utilities can leverage back-office excellence as a strategic advantage, the role of advanced analytics in decarbonization efforts, and why the most successful energy companies are those that recognize transformation as both a technological and operational imperative.
How are traditional utilities balancing the urgent need for decentralization with the reliability requirements that have defined the industry for decades, and what role do back-office systems play in this transformation?
Kunal Saxena: The pace of technological innovation especially with the surge of AI and cloud computing infrastructure has triggered an unprecedented demand for utilities. This comes alongside an ongoing push to decentralize distribution to increase reliability and reduce operations cost. DERs are not just the solution, but also a strategic advantage when it comes to optimization.
Several traditional utilities I have worked with have implemented or are in the process of grid modernization programs that are supported by rate cases. The rate cases for public utilities often highlight the need to increase energy prices on the consumer in order to recoup the cost of implementing DERs in the network. For privately held utilities this is relatively easier, however, the financial value determination and the rate changes passed on to consumers need to strike a delicate balance. In fact, simply put, decentralization isn’t the future anymore, it’s here.
However, It is a highly analytical and dynamic challenge to offset the direct investment needed to implement advance metering and bi-directional transmissions systems by simply sharing the cost with consumers. The challenge now is ‘How does a Utility derive value from it?’. This is where the back office comes in. With a bidirectional grid, the scenarios for optimizing costs to consumers and recovering cost of modernization while deriving more and more energy from renewable sources become more complex. The back office systems that used to support a centralized grid model instantly become handicapped and lack the scalability to support these complex financial models/scenarios. The challenges to scale traditional on-premise back office with grid modernization was crippling a major Generations company in the northeast. The cost to modernize was upwards of $100 M however, the cost to scale existing back office apps was significantly higher ~$160M. The answer was simple: optimize back office processes , adopt nimble, cloud-enabled solutions to power the future and streamline the ongoing resourcing / operations cost. In another case, a major energy utilities provider in the Mid-west modernized its back office for about $60M reaping ~$100M in savings over 5 years while enabling higher scalability.
In your experience advising C-Suite executives on technology innovation, what are the most critical misconceptions energy leaders have about implementing AI and cloud technologies in utility operations?
Kunal Saxena: The C-suite faces a variety of challenges as they look ahead into the next era of generations and distribution. Generation industry has changed with hyper focus on renewables and carbon neutrality, whereas distribution is looking at potential overhaul with modern meters, bi directional grids, drones and many more technologies. This dramatic shift has led to a proliferation of vendors loosely using hot terms like ‘AI’ and ‘Cloud’ to grab executive attention.
The CIO of a major renewable energy generation company based out of California expressed his concerns about how AI could be differentiated from traditional automation, the CP of operations mentioned along the same lines, how many of these theoretical use cases are actually solved by AI in reality? More commonly the theme is ‘has this been done before successfully?’ They have plenty of use cases where they could see the applicability of AI like ‘predictive maintenance’, ‘virtual agent enabled procurement’, ‘proactive intelligent forecasting’ and the list goes on. However, when they reach out they hear different answers from different vendors and tools. This leads to misconceptions about what AI ‘Can’ and ‘Cannot’ do.
The second and a rather important challenge that C-Suite struggles with is the contextualization of data security within the cloud and within AI. AI is enabled as another layer over the cloud infrastructure and it takes some thoughtful discussions to analyze the data governance and security model that protect the organization data. A good vendor partner is one that delineates the two and helps the C-suite understand both layers of data security paradigms independently and in tandem.
You’ve written pretty extensively about back-office excellence being the hidden driver of utility modernization. Can you share a specific example where optimizing financial processes delivered unexpected operational benefits for an energy company?
Kunal Saxena: The goal for back off modernization is truly to enable operations efficiency. However, through these transformation journeys enterprises often uncover drivers of value that were otherwise nondescript. A running theme in deriving hidden value is through establishing a common process language. Utilities have a wide process footprint that goes well beyond finance. Field services, maintenance, inventory operations, taxation are all impacted by a finance back office transformation. What many clients fail to realize is that by way of streamlining finance functions the enterprise adopts a consistent language.
This brings with it, clean data, entity definitions, accounting and reporting hygiene and thereby driving heightened value at the operations end where engineers inputting work order spend less key strokes and secure faster approvals for equipment, they are able to dispatch crew faster given clearer visibility into personnel schedule and utilization . Predictive maintenance and proactive stocking is possible when the enterprise cleanses itself of the garbage data that clouds the operations efficacy. In such cases, the field engineers don’t even realize that they have fewer projects to choose from, as the security is contextual thereby making their choices in transaction entry and functions easier and intelligent.
How do you approach the challenge of integrating modern cloud-enabled platforms with legacy utility systems that were built for centralized operations, particularly when M&A activities are involved?
Kunal Saxena: Modern cloud platforms have brought with them the concept of a technology agnostic middleware that ingests data from disparate sources in a variety of formats allowing the SaaS applications and the eventual user experience to be seamless and consistent. When I design the underlying architecture supporting the ERP footprint, I start with a multi-pillar technology strategy which addresses the key dimensions of the enterprise : Data Strategy, Infrastructure Strategy, Security Strategy and Application Integration Strategy.
I investigate the technology footprint with different applicable dimensions for each of these pillars to arrive at a plan that meets the program goals. One key component in all of these pillars is the current / legacy application footprint. I analyse the future state design across these 4 pillars to comprehensively integrate only the required applications in the future state with the necessary data and security constraints enabled over the most suitable infrastructure.
M&A activities provide a springboard for transformations allowing organizations entering the transaction to use this event as an opportunity to streamline processes and accelerate innovation. In one of the most comprehensive overhauls of the last decade in the Utilities industry a major enterprise moved from on premise legacy systems to cloud enabled platforms. This sweeping change was a multi-year effort that required hundreds of resources to work tirelessly through a major M&A event to achieve the enterprise goals. I applied the same approach of a multi-pillar strategy to design their future state solution architecture with both the entities entering into the M&A transaction in mind. The analysis resulted in a scalable, modern, cloud enabled solution that embraced the unique processes applicable to each organization while harmonizing similarities to drive efficiencies.
What are the biggest technical and financial obstacles utilities face when implementing decarbonization strategies, and how can advanced analytics help overcome these barriers?
Kunal Saxena: The complexities associated with decarbonization impact all areas of utilities operations. From implementation of smart meters to grid modernization, to advanced software to manage bi-directional energy flows and transfers, tracking associated billing events to balance costs against renewable energy credits, the effort to decarbonize truly requires a transformation of the enterprise. Each of these elements of implementing decarbonization can be assessed through the lens of cost to implement, speed to implement, value derived, and strategic alignment with the enterprise. While evaluating challenges the most important aspect is the enterprise context.
The same initiative could have a value that is exceptional for one enterprise, and not a consideration for another. Since each company is at a unique position with their user base, technologies and processes, the idea of prioritizing transformation challenges is hard to fathom. Depending on the business context, the challenges complexity in terms of the aforementioned factors can vary thereby impacting how an enterprise sees them.
The running theme I notice in Utilities is that there is a lack of vision to unlock the true potential using data stored in these smart devices. Enterprises continue to operate in silos where operations often highlight tactical challenges to adoption of new equipment whereas finance struggles to think beyond the metrics generated. A common technical challenge is the truly harmonized integration between the modern hardware that’s generating significantly higher quality data, but the back office is not capable of ingesting/processing such data to provide the right analytics or insights needed for strategic decision making. Corporations must truly strive to understand the operational metrics as much as the operations should align with financial goals of the organization. For Example Finance establishes KPIs that require analytics based on the limited exposure to metering/usage/timing capabilities of a smart grid, however Operations continues to focus on outdated metrics needed for scheduling maintenance, resource and equipment forecasting. The right level of these metrics may not be in alignment due to outdated back office systems if there is a lack of seamless data exchange between the operational and financial systems.
In such cases modern analytics systems can provide significant value. Today’s analytics systems powered by AI do not need significant development effort to help bridge the gap between Operations data and Enterprise level metrics. A modern top-tier analytical solution can be plugged in over the legacy enterprise to bridge this gap, however, to truly gain the most value out of analytics, the back office systems need to be harmonized to receive a common level of information driven by a standardized data set.
As utilities increasingly adopt distributed energy resources and smart grid technologies, how should they restructure their enterprise systems to handle the exponential increase in data and decision points?
Kunal Saxena: I use a framework for helping utilities achieve their path to back office modernization. I break the enterprise systems to align with key end to end business processes and then break those down to line up with prioritized components between end to end processes. This can be visualized as a heat map that overlays the interconnected nature of enterprise systems with their prioritized process. Many enterprises know a cold listing of the applications they have, but they fail to recognize the interdependence within applications without such a visual representation.
I then use this visual to help prioritize the ongoing enterprise programs (like ongoing initiatives, strategic upgrades). This is then extended to build out a resource capability matrix that outlays current resources and future resource demands against these initiatives. Now, the executives can see their organization’s readiness to adapt to these changes. However, resources and priorities are not sufficient, the next dimension needed in this analysis is the value ($$) associated with the process – like operational cost , estimated benefits or hours of effort saved through the program.
With this mutli-dimensional view of their existing enterprise, the executives can now specifically understand which areas or processes are under-served and which ones are costing the enterprise a premium. From this point on the executives feel empowered to make the right decision about their enterprise systems – whether to upgrade, hold, decommission or outsource.
This approach brings into perspective the modern capabilities of handling multitudes of data elements and decision points generated by modern sensors, meters, and ties it to the specific areas of enterprise that need improvement.
Looking ahead 3-5 years, how do you see Generative AI transforming utility operations beyond customer service applications, particularly in areas like predictive maintenance and grid optimization?
Kunal Saxena: I have been spending a lot of time on generative AI, particularly from modern LLMs like ChatGPT, Perplexity, Gemini, Grok and other similar tools/solutions. It is clear to me that the future of problem solving involves these large scale LLMs to a great degree. These solutions are more targeted towards corporate back office processes given their conversational nature.
In operations, though, I see a higher likelihood of agentic AI usage. Three to five years from now, all enterprises would be using powerful, but specific agents, potentially from more than one vendor to automate day-to-day functions. Procurement, inventory movements, task assignments for maintenance and field services, technician dispatch, inventory replenishments and many day to day functions can be automated with AI agents that constantly receive feedback from other AI solutions that provide information on proactive, predictive maintenance. E.g. weather watcher AI could help prepare for updates ahead of a major storm event by automating procurement and inventory transfers. A LLM being used in the HQ could signal the issuance peak of energy usage thereby highlighting technician availability concerns to the field without the corporate having to send a single email.
Beyond LLMs, AI could be embedded in the hardware itself making the grid modulate its components based on energy demands. AI agents in trucks could automate the dispatch and routing during a major storm or natural disaster thereby providing safe dispatching and transfer of people and equipment. There are a plethora of use cases, however, the capabilities need to be tested significantly. The growing concern with AI agents is their accuracy for critical use cases (e.g. Storm relief, grid control). Some areas where I foresee slower adoption would be those tied to nuclear energy where ‘human intervention and oversight isn’t just critical, it is legally mandated’
Overall, AI adoption will increase many-fold in the next five years, which will further strengthen the need for robust enterprise infrastructure and applications.
What advice would you give to energy executives who are hesitant about investing in modern back-office transformation due to concerns about regulatory compliance and operational disruption?
Kunal Saxena: Executives from different businesses have shared a multitude of concerns with me in the last few years. Regulatory compliance is actually a driver for many of these executives to move to a modern back office solution. As utilities are on a growth curve, many of them struggle to balance GAAP and FERC needs without a modern ERP. More so taxation also can get tricky if the back office systems are not capable of automating it. However, some concerns also stem from the complex nature of utilities business which isn’t inherently supported by out of the box applications. Executives fear that even after investing millions of dollars in a modern back office system, they will end up sacrificing functionality they currently have.
I always try to anchor them back to their goals, and that sometimes ‘Do Nothing’ is not an option at all. I highlight to them their competitors that have already embarked on similar journeys and some that have established competitive advantage with a modern enterprise. In today’s transformation ‘downtime’ is a monomer and the decades old idea of operational disruption has been addressed through phased roll-outs, continuous upgrade models. I always ask them back, what is the cost to your organization for ‘Doing nothing’.
Overall, most executives are already on board with the idea of modernizing their systems, their struggles are more around dropping a $$ tag to the value this modernization effort brings to them. In such conversations I advise them to perform that internal value determination exercise and confirm that the goals they seek to achieve are ‘worth the effort’.
Drawing from your 15 years across energy, oil & gas, and nuclear sectors, which technological innovations from other industries do you believe utilities are overlooking that could dramatically improve their operational efficiency?
Kunal Saxena: The Oil & Gas sector has seen tremendous growth by investing in modern back office technologies. The most notable difference I notice is that their business processes are streamlined helping them achieve higher value in reporting and analytics solutions.
As I look at healthcare, they also have to navigate major data security and compliance challenges, however, most healthcare companies have refined and highly complex billing solutions in place with modern enterprise systems and they are able to navigate the M&A landscape with more agility.
The banking sector has been at the forefront of technology adoption with increased focus on Gen AI and advanced financial models to help with their strategies.
Utilities stand at the cusp of this shifting landscape where they can absorb the advanced reporting and analytics capabilities seen in banking, the agility to navigate M&A activities in the healthcare sector by way of implementing scalable data models and enterprise applications and adopting a cloud-first approach to any system they implement.
How can utility companies leverage the scalable enterprise platforms you’ve implemented to not only optimize current operations but also position themselves as technology leaders in the broader energy ecosystem?
Kunal Saxena: Over the last 19 years I have led, advised on and assessed the implementation of best of breed Enterprise Systems from vendors like Oracle, Salesforce, SAP and Workday. A cloud first approach supplemented by a value driven prioritized plan has always worked for the clients that vary from fortune 50 to Fortune 1000. Any enterprise, large or small, needs stability, scalability and most of all visibility into the future of its internal operations to develop a growth mindset. This in-office stability comes from an application footprint that can sustain several M&A events, growth and recessions cycles, unanticipated macroeconomic factors and legal/mandated changes. A truly cloud enabled enterprise needs to be able to adapt and scale Up Or Down to meet the market needs, so should its systems. A SaaS/Cloud enabled enterprise therefore takes center stage in this regard.
The second tenet is to keep an agile mindset. Large scale, long duration programs tend to cripple the organization’s ability, therefore the focus should be on speed rather than bespoke tailored solutioning.
Utilities can adopt a streamlined leading process driven enterprise solution that enables 80% of their operations with minimal modifications, to get onboard the Cloud platform, This solution not only simplifies the application footprint, but also reduces the ongoing operations and maintenance overhead. With this efficiency, and given that most of the data is aligned to one standard, the organization is able to move faster and get real-time data needed to make strategic decisions. Utility can leverage big data, advanced analytics, AI powered agents, and in near future, agentic AI itself to further reduce error prone manual processes.
A utility that is struggling with inventory management, can use such a system to optimize inventory and better negotiate supplier contracts. A utility that hasn’t been making accurate forecasts can now use real-time data from all its applications to accurately update its forecast based on right-on-time actuals.
A utility can now use the most accurate transactional data to support its rate case. Utilities can predict, and proactively update their resources (labor and non labor) to meet weather driven, economic, and other fluctuations in demand and supply.
Needless to say all of this information is available faster, and with a higher degree of accuracy all driven by the core enterprise systems.