Why Utilities should evolve from reactive collections to Intelligent revenue operations

The obsolescence of the reactive, Low-Empathy model 

The utility market navigates an environment of high volatility where geopolitical tensions and rising energy costs collide with a severe affordability crisis. In the United Kingdom, for instance, the domestic energy total debt climbed to a record £4.55 billion by the end of 2025, representing an 18% year-on-year increase. From the onset of the energy crisis, average household electricity arrears rose by 35% to £2,036, while average gas arrears surged by 44% to £1,610 [1].  

While corporate debt management strategies attempt to improve overall metrics, the days sales outstanding (DSO) has steadily increased across the EU, 1.8% over the past decade to an average of 50 days [2]. Furthermore, Western Europe recorded a four-day surge in corporate Working Capital Requirements (WCR) for the third consecutive year [3]. This growing duration of delinquency forces utilities to adopt defensive debt management strategies to shield their balance sheets, which ultimately inflates the utility’s regulated revenue requirement. For instance, debt relief schemes that write off historic arrears often fund this relief by socializing costs across all bill-payers, adding an annual surcharge that risks pushing more marginal consumers into delinquency [4]. 

When faced with this panorama, it is evident that the traditional model is no longer sufficient. Managing debt only after its due date is a reactive measure that kicks in when the financial damage is already done. By applying rigid, mass collection strategies, this model shows a total disconnect from the user’s reality. By failing to mix data intelligence with empathy, organizations ignore the macroeconomic pressures facing their customers, such as the 38% of Scottish households forced to ration heating [5] and push them into delinquency instead of offering preventive solutions. The true problem escapes conventional indicators, as revenue leakage begins long before the customer reaches the collections stage.  

The real problem: Revenue is lost before collections 

Most utilities approach revenue recovery through overdue accounts. However, a significant portion of revenue loss occurs much earlier in the commercial cycle through metering errors, tariff inconsistencies, delayed billing, reconciliation issues, and operational anomalies that remain undetected until it is too late. 

The financial scale of these pre-collection losses is massive. Across Europe, an average of 25% of all treated water is lost as non-revenue water (NRW) due to commercial and physical errors before it can generate revenue [6]. In the UK alone, leakage and measurement discrepancies account for the equivalent of 1,245 Olympic swimming pools of water lost every single day, severely draining potential cash flow [7]. What is most concerning is that these losses rarely appear on standard collection reports. By the time the organization detects the problem, much of the potential revenue has already been diluted throughout the commercial cycle. 

All of that is happening without putting in the equation the constraints and realities of each household. When systems  can’t dynamically review demographic and socioeconomic variables, utilities lose the opportunity to act with empathy. True financial retention occurs when the system, prior to default, can proactively segment customers based on financial vulnerability profiles, for example. Utilizing data intelligence to automatically offer flexible payment plans, social tariff enrollments, or dunning deferrals to vulnerable customers maintains collection rates while protecting the customer relationship. This can prevent families from being pushed into uncollectible debt by the rigidity of the traditional model. 

The hidden cost of fragmented architectures and legacy systems 

It is common to find utilities where billing, customer management, advanced metering, CX, and field work systems operate independently, forcing teams to work with scattered information and manual processes. When measurement, consumption, billing, and payment data are not synchronized in real time, concrete problems emerge like unbilled consumption, tariff errors, payment application failures, delayed service orders, and commercial disputes that stall collections.  

This chain of failures is known as revenue leakage, the economic value progressively lost throughout the commercial cycle before being detected by financial departments. The gaps in debt recovery are directly tied to critical technological deficiencies. According to Gartner, traditional CIS platforms were built as monolithic, batch-processing engines optimized for simple meter-to-cash cycles, and they lack the architectural flexibility needed to support complex data flows and dynamic pricing [8]. Smart meter data arrives late to commercial processes, consumption anomalies are processed weeks later, and the result is a growing DSO, compressed margins, and a progressively deteriorating customer experience.  

When Volume Exceeds Human Capacity  

The energy transition has exponentially multiplied the amount of data generated by advanced metering infrastructure. A few years back, utilities had already installed over 120 million smart metering devices [9]. However, many utilities still rely heavily on manual exception queues and spreadsheet-based tracking to identify financial problems. The gap between the available data volume and the capacity to process it manually becomes increasingly difficult to close.  

A recent industry case illustrates the magnitude of this challenge: a utility with approximately 480,000 smart meters identified more than 95,000 leakage alerts in just 72 hours using advanced analytics and artificial intelligence [10]. That level of detection and exception management is simply not viable with manual processes. Meanwhile, when billing errors, estimated invoices, or confusing rate structures occur, contact centers are flooded with inquiries. Resolving these issues manually requires extensive call center staffing, driving up transaction handle times and operational overhead. Waiting until the end of the billing cycle to identify anomalies means allowing losses to accumulate for weeks before initiating any corrective action.  

The solution: Intelligent revenue operations  

Faced with this landscape, more utilities are moving toward an Intelligent revenue operations model. Unlike traditional collections approaches, this strategy does not wait until revenue is lost before taking action. Instead, it uses predictive analytics, automation, and artificial intelligence to identify and address risks throughout the commercial cycle before they become financial losses. Moving to this model requires utilities to rethink how data, customer operations, field activities, billing, customer experience, and financial management work together across the organization.