Greece will introduce revised balancing-market rules on November 1, giving power suppliers a stronger financial incentive to forecast customer consumption accurately. Energy regulator RAAEY confirmed the start date after transmission operator IPTO established that the required technical systems were ready. The provisions address significant, systematic differences between the demand market participants declare and the electricity their customers ultimately use.
The rules are not intended to penalise every forecasting mistake: electricity demand cannot be predicted perfectly. Weather changes, industrial output varies and individual customers continually alter their behaviour. The regulatory focus is on significant and systematic deviations that show a portfolio is repeatedly scheduled inaccurately.
Persistent errors can impose costs on the wider system. When a supplier regularly declares less demand than its customers consume, the transmission operator must secure additional electricity in the balancing market. Repeated overestimates may instead require generation elsewhere to be reduced; either pattern can increase balancing requirements.
Forecasting and supplier margins
From November, weak forecasting will carry a more explicit financial consequence for electricity retailers. Suppliers have traditionally competed on prices, discounts, contract terms and customer acquisition, but forecasting quality will also affect margins. Two suppliers purchasing power at similar wholesale prices may achieve different profitability if one repeatedly incurs greater imbalance costs, giving the more accurate forecaster a structural advantage.
That incentive creates an opportunity for forecasting-as-a-service. Specialist software providers can use historical meter readings alongside weather forecasts, industrial schedules, calendar effects, customer characteristics and real-time consumption information to estimate portfolio demand. Machine-learning models can be retrained continuously as customer behaviour changes.
Such tools are particularly relevant to suppliers whose portfolios are changing rapidly. A retailer taking on thousands of customers cannot assume their consumption resembles that of its existing customers: homes with electric heating differ from those using conventional heating, hotels respond strongly to tourism patterns, and industrial users follow production schedules. EV charging adds a further, increasingly variable load, making forecasts more customer-specific.
Meter data and flexible demand
Smart-meter data can materially improve forecast accuracy because monthly readings reveal little about when electricity is used. Interval meters show consumption by hour or quarter-hour, helping suppliers distinguish predictable peaks from volatile loads. Beyond billing, those records allow retailers to build more detailed models of their exposure to balancing costs.
The data can also support aggregation and demand response. A supplier that knows a customer routinely uses 5 MW during a particular production period can forecast that load more accurately. If some of it is flexible, an aggregator may also be able to change the consumption itself, bringing forecasting and flexibility closer together.
An advanced portfolio manager could therefore do more than identify a likely imbalance: it could correct it before delivery. A supplier expecting demand above its nominated position could buy electricity intraday or ask participating flexible customers to reduce use. If demand is likely to fall short of its forecast, controllable loads, including EV charging or thermal systems, could potentially be increased.
Active portfolio management creates opportunities for automated intraday trading and optimisation providers. A retail platform could repeatedly update its demand forecast, compare it with existing market positions and execute corrective trades as delivery approaches. Human traders would increasingly supervise those systems rather than calculate every adjustment manually.
A more complex retail market
The rule change comes as Greece’s retail market becomes harder to forecast using conventional historical averages. Customers are increasingly exposed to fixed, variable and dynamic products, while distributed solar changes their net consumption. Smart meters supply more detailed data, and electric vehicles and behind-the-meter equipment introduce new load patterns. Data centres and other large consumers add further concentrations of demand.
As balancing markets become more sophisticated, the cost of an inaccurate forecast also becomes more important. That could affect how electricity-supply businesses are valued: customer numbers alone may tell an investor less about a retailer’s risks. Two suppliers with 500,000 customers each can have markedly different profiles because of customer mix, data quality, forecasting technology and balancing performance. A high-quality consumption dataset or a proven forecasting model could become an intangible asset.
Industrial customers may also see commercial effects. Suppliers may increasingly favour companies that provide accurate operating schedules or access to real-time consumption data, while customers with unpredictable demand create balancing costs. Stable or transparent consumption profiles could consequently become more attractive to retailers, and contracts could eventually reflect the difference.
Contracts and aggregation
Energy-intensive customers may be offered better terms if they share production forecasts, permit automated access to meter data or participate in arrangements that help suppliers manage deviations. Such arrangements would further narrow the boundary between electricity supply and energy management. They would also give suppliers more information and options for managing customer demand.
Independent aggregators could benefit if the financial consequences of systematic forecasting errors increase suppliers’ demand for flexible resources. Aggregators controlling refrigeration, HVAC, EV charging, industrial processes or distributed generation can provide capacity to correct portfolio positions. Forecasting a deviation and then addressing it could therefore create a potential market for flexibility.
The revised rules will not eliminate imbalance costs, and balancing markets remain necessary because every forecast contains some error. Their purpose is to discourage persistent behaviour that places avoidable costs on the electricity system. Software that reduces systematic portfolio errors can protect supplier margins, while smart-meter data, automated trading and flexible customers become more valuable to retailers. A supplier’s competitive position may increasingly depend not only on its wholesale purchase price but on how accurately it predicts consumption, adjusts its position before delivery and limits the cost of remaining errors.
