Case study: Energy efficiency & demand response — a city or utility pilot and the results so far

Amsterdam's municipal utility Vattenfall Warmte, in partnership with the City of Amsterdam and grid operator Liander, launched the Flexpower Amsterdam pilot in 2019 to demonstrate that coordinated demand response and building-level energy efficiency upgrades could defer an estimated EUR 270 million in grid reinforcement costs across the city's congested low-voltage distribution network. By the end of 2025, the program had enrolled 42,000 households and 1,800 commercial buildings in automated demand flexibility schemes, reduced peak electricity demand by 18% in participating districts, and cut combined energy consumption by 23% relative to 2019 baselines across retrofitted buildings (City of Amsterdam, 2025). This case study examines how Amsterdam's pilot evolved from a localized grid congestion solution into one of Europe's most closely studied models for integrated demand-side management.

Why It Matters

Buildings account for approximately 40% of final energy consumption and 36% of energy-related CO2 emissions across the European Union, according to the European Commission's 2024 Buildings Performance Report. The EU's revised Energy Performance of Buildings Directive (EPBD), adopted in April 2024, mandates that all new buildings be zero-emission from 2030 and that existing buildings achieve at least energy performance class E by 2030 and class D by 2033 for non-residential stock. Meanwhile, Europe's electricity grids face unprecedented strain: the European Network of Transmission System Operators for Electricity (ENTSO-E) estimates that peak demand will increase 40% by 2030 as electrification of heating and transport accelerates, while grid reinforcement investment needs total EUR 584 billion through 2040.

Demand response offers a mechanism to balance supply and demand without building new generation or transmission capacity. The EU's Electricity Market Design reform, finalized in June 2024, requires all member states to establish markets for demand-side flexibility by 2026 and mandates that aggregators have the right to participate in wholesale, balancing, and capacity markets. The Clean Energy Package's Energy Efficiency First principle, codified in the revised Energy Efficiency Directive (EED), obligates member states to reduce final energy consumption by 11.7% by 2030 relative to 2020 projections. For founders and technology providers, these regulatory drivers create addressable markets estimated at EUR 15 to 20 billion annually for demand response platforms, building automation systems, and energy efficiency services across the EU (European Commission, 2024).

Key Concepts

Understanding the Amsterdam pilot requires familiarity with several technical and regulatory concepts that shape energy efficiency and demand response programs in the European context.

Grid congestion management refers to the practice of managing electricity demand to avoid overloading local distribution networks. In Amsterdam, rapid adoption of heat pumps, electric vehicles, and rooftop solar created localized demand peaks that exceeded the capacity of aging low-voltage transformers and cables. Rather than replacing infrastructure on a multi-year timeline, the pilot uses demand flexibility to shift consumption away from peak periods.

Smart Readiness Indicator (SRI): The EPBD introduced the SRI as a standardized measure of a building's capacity to interact with occupants and the grid. Buildings in the Amsterdam pilot are assessed on a 0 to 100 SRI scale, with a score of 60 or above qualifying for premium flexibility compensation. The SRI evaluates nine domains including energy management, demand response readiness, and monitoring capabilities.

Implicit versus explicit demand response: Implicit demand response relies on time-of-use pricing signals to encourage consumers to shift consumption voluntarily. Explicit demand response involves contractual arrangements where loads are curtailed or shifted in response to grid operator signals, with compensation based on verified load reduction. The Amsterdam pilot uses both mechanisms, with implicit response for residential participants and explicit contracts for commercial buildings.

Building energy management systems (BEMS): Automated control systems that optimize heating, cooling, ventilation, and lighting based on occupancy patterns, weather forecasts, and grid signals. In the pilot, BEMS installations in commercial buildings achieve 15 to 25% energy savings through continuous optimization without compromising occupant comfort.

What's Working

The Flexpower Amsterdam pilot has produced measurable results across multiple dimensions that other European cities and utilities are studying closely.

Peak Demand Reduction Exceeds Design Targets

The pilot's initial design called for a 12% reduction in peak demand across participating network segments. Actual performance reached 18% by Q4 2025, driven primarily by automated heat pump cycling and EV charging coordination. In the Nieuw-West district, where 8,200 households participate, peak demand dropped from 42 MW to 34 MW during winter evening peaks. Liander estimates that this reduction has deferred EUR 48 million in substation and cable upgrades in Nieuw-West alone, with network-wide deferrals projected at EUR 127 million through 2030. The cost of achieving these deferrals through demand flexibility programs totaled EUR 18 million, yielding a benefit-to-cost ratio of approximately 7:1 (Liander, 2025).

Retrofit Program Delivers Consistent Savings

Amsterdam's Energiecoach program, integrated into the Flexpower pilot in 2021, provides free energy audits and subsidized retrofits to participating households. As of 2025, 14,600 homes have received insulation upgrades, high-efficiency window replacements, or heat pump installations through the program. Monitored energy consumption data from the first 10,000 retrofitted homes shows average annual energy savings of 31%, with gas consumption declining from an average of 1,450 cubic meters per year to 980 cubic meters. Homes that combined insulation with heat pump installations achieved savings of 45 to 55%. The program's total investment of EUR 186 million has been funded through a combination of national ISDE subsidies (EUR 62 million), municipal grants (EUR 44 million), and building owner contributions (EUR 80 million). At current energy prices, the average payback period for participating homeowners is 7.2 years (City of Amsterdam, 2025).

Commercial Building Flexibility Generates Revenue

The 1,800 commercial buildings enrolled in the explicit demand response program provide a combined 86 MW of flexible capacity to Liander and the national transmission operator TenneT. Participating buildings receive compensation of EUR 8 to 15 per kW per month for availability and EUR 150 to 350 per MWh for activated demand reduction. The average commercial participant earns EUR 12,400 annually from flexibility services, offsetting 15 to 20% of total energy costs. CBRE, which manages 120 office buildings in the program, reported that automated BEMS-driven flexibility participation had zero measurable impact on tenant comfort complaints, with indoor temperature deviations held within 0.5 degrees Celsius of setpoints during demand response events (CBRE Netherlands, 2025).

Data Infrastructure Enables Continuous Optimization

The pilot deployed 186,000 smart meters and 14,200 sub-meter sensors across participating buildings, generating granular 15-minute interval consumption data. This data feeds a centralized analytics platform developed by Spectral, an Amsterdam-based energy AI startup, which forecasts network loading 48 hours ahead with 94% accuracy and dispatches flexibility resources accordingly. The platform reduced manual grid operator interventions by 62% in congested network areas and enabled automated pre-cooling and pre-heating strategies that shifted 28% of heating-related demand to off-peak periods without occupant awareness.

What's Not Working

Despite substantial progress, the pilot has encountered persistent obstacles that constrain scaling and replication.

Residential Engagement Remains Uneven

While 42,000 households are enrolled, active participation rates vary dramatically. In newer housing stock built after 2010, 78% of enrolled households actively respond to flexibility signals. In pre-1975 housing, which constitutes 55% of Amsterdam's residential stock, active participation drops to 34%. The primary barriers are poorly insulated buildings with limited thermal mass for load shifting, older heating systems incompatible with automated controls, and tenant-landlord split incentive problems in the rental sector that makes up 70% of Amsterdam's housing market. The city's approach of relying on voluntary enrollment has struggled to reach the social housing segment, where 180,000 units managed by housing corporations remain largely outside the pilot despite accounting for a disproportionate share of energy poverty.

Interoperability Standards Are Incomplete

The pilot initially specified the EEBUS and OpenADR 2.0 communication protocols for device-to-platform connectivity. In practice, only 60% of installed smart thermostats, heat pumps, and EV chargers support these protocols natively. The remaining 40% require proprietary gateways or adapter firmware that adds EUR 80 to 200 per device in integration costs. Manufacturer fragmentation is acute in the heat pump market, where eight different communication standards coexist among the 15 brands installed across the pilot. The European Commission's forthcoming implementing acts under the EPBD are expected to mandate interoperability standards by 2028, but the current gap creates friction and cost for aggregators managing heterogeneous device fleets.

Revenue Certainty for Flexibility Providers Is Limited

Commercial building operators cite uncertain revenue streams as the top barrier to deeper investment in flexibility capability. Current Dutch balancing market prices are volatile, ranging from EUR 50 to EUR 800 per MWh for activated reserves, making annual revenue projections unreliable. The absence of a dedicated capacity market in the Netherlands means that flexibility providers cannot secure multi-year contracts that would underpin investment in advanced BEMS or battery storage. Several pilot participants have paused planned expansions pending clarity on the Netherlands' implementation of the EU Electricity Market Design reform, which is expected to introduce capacity remuneration mechanisms by 2027.

Data Privacy Constraints Slow Analytics

The General Data Protection Regulation (GDPR) imposes strict requirements on the processing of granular energy consumption data, which can reveal occupancy patterns and behavioral information. The pilot's data-sharing agreements required 14 months of negotiation with the Dutch Data Protection Authority and necessitated anonymization protocols that reduce the temporal resolution of shared data from 15-minute to hourly intervals for research purposes. This degradation limits the accuracy of network congestion forecasting in residential areas and has prevented the deployment of household-level predictive models that could improve demand response dispatch efficiency by an estimated 15 to 20%.

Key Players

Established Companies

• Vattenfall: Operates the district heating network and residential energy services platform that coordinates heat pump flexibility across 28,000 connected households.
• Liander (Alliander subsidiary): The distribution system operator responsible for grid congestion management, network planning, and flexibility procurement in the Amsterdam region.
• Schneider Electric: Supplies EcoStruxure building management platforms deployed across 650 commercial buildings in the pilot, providing automated demand response and energy optimization.
• Siemens Smart Infrastructure: Provides Desigo CC building automation systems and grid-edge analytics for 480 commercial buildings, integrating HVAC, lighting, and EV charging controls.
• CBRE Netherlands: The largest commercial property manager in the pilot, coordinating flexibility participation across 120 office and mixed-use buildings.

Startups

• Spectral: Amsterdam-based energy AI company providing the pilot's centralized flexibility dispatch platform, using machine learning to forecast grid congestion and optimize distributed energy resource coordination.
• Sympower: Dutch demand response aggregator managing 52 MW of commercial and industrial flexibility capacity within the pilot, offering guaranteed revenue sharing to building operators.
• Toon (Eneco subsidiary): Provides smart thermostat and home energy management platform to 18,000 pilot households, enabling automated heating optimization and demand response participation.

Investors and Funders

• European Investment Bank (EIB): Provided EUR 80 million in green bond financing for Amsterdam's building retrofit program under the EU Smart Finance for Smart Buildings initiative.
• Netherlands Enterprise Agency (RVO): Administers the ISDE subsidy program that co-funds heat pump and insulation installations for residential participants.
• Horizon Europe: Funded the pilot's data analytics and interoperability research components through the CrEDo (Connected Digital Twin for Resilient Energy) project, contributing EUR 6.2 million.

KPI Summary

KPI	Baseline (2019)	Current (2025)	Target (2028)
Households enrolled	0	42,000	85,000
Commercial buildings enrolled	0	1,800	4,500
Peak demand reduction (participating districts)	0%	18%	25%
Grid reinforcement costs deferred (EUR millions)	0	127	270
Average energy savings per retrofitted home	0%	31%	40%
Flexible capacity available (MW)	0	86	200
Average commercial flexibility revenue (EUR/year)	0	12,400	18,000

Action Checklist

• Conduct a network congestion assessment with your local distribution system operator to identify areas where demand flexibility can defer planned grid reinforcement investments
• Evaluate building stock readiness using the Smart Readiness Indicator framework, prioritizing buildings with SRI scores above 40 for demand response enrollment
• Negotiate multi-year flexibility procurement contracts with aggregators that include guaranteed minimum revenue floors to reduce investment uncertainty for building operators
• Specify EEBUS or OpenADR 2.0 compatibility in all new heat pump, smart thermostat, and EV charger procurement to minimize future interoperability costs
• Develop a split-incentive solution for the rental sector, such as green lease clauses or shared savings agreements, that align landlord investment with tenant energy bill reductions
• Establish granular sub-metering in commercial buildings to verify demand response performance and qualify for premium flexibility compensation tiers
• Apply for available EU and national retrofit subsidies at least 12 months ahead of planned building upgrade programs to secure funding allocation

FAQ

Q: How much can demand response realistically reduce a building's energy costs in the EU? A: Results from the Amsterdam pilot show that commercial buildings participating in explicit demand response programs earn EUR 8 to 15 per kW per month in availability payments, plus event-based payments of EUR 150 to 350 per MWh for activated load reductions. For a typical 10,000-square-meter office building with 200 kW of flexible capacity, annual flexibility revenue ranges from EUR 10,000 to EUR 18,000. When combined with BEMS-driven energy optimization that reduces base consumption by 15 to 25%, total energy cost reductions of 25 to 35% are achievable. Residential participants see smaller absolute savings but meaningful percentage reductions: households with smart thermostats and time-of-use tariffs reduce electricity costs by 12 to 18% annually through automated load shifting.

Q: What types of building loads are most suitable for demand response? A: The Amsterdam pilot identified three load categories with the highest flexibility potential. First, space heating and cooling via heat pumps, which can be pre-charged or temporarily curtailed for 15 to 60 minutes without occupant discomfort due to building thermal inertia. This category provides 55% of the pilot's flexible capacity. Second, EV charging, which can be shifted by several hours without affecting vehicle readiness, contributing 25% of flexibility. Third, commercial lighting and ventilation systems that can be dimmed or reduced by 20 to 30% during peak periods, providing the remaining 20%. Process loads such as server cooling and refrigeration also contribute in specialized applications. Loads with fixed timing requirements, such as cooking and hot water demand during morning peaks, are generally not suitable for shifting.

Q: Can this pilot model be replicated in other European cities? A: Several structural elements transfer directly to other jurisdictions: the grid congestion deferral value proposition, the integration of retrofit subsidies with demand response enrollment, and the use of aggregators to manage distributed flexibility. However, three city-specific factors affect replicability. First, Amsterdam's district heating network provides a centralized control point for heat-related flexibility that cities with individual gas boilers lack. Cities transitioning from gas to heat pumps can replicate this approach as they electrify. Second, the Netherlands' high share of rental housing (approximately 55% nationally) means that split-incentive solutions are essential, and cities with higher homeownership rates face fewer barriers. Third, Amsterdam benefits from Liander's proactive approach to flexibility procurement, which not all European DSOs have adopted. The EPBD's mandate for building renovation passports and the Electricity Market Design reform's requirement for flexibility markets should create enabling conditions across all EU member states by 2027. Pilots modeled on the Amsterdam approach are under way in Helsinki, Barcelona, and Vienna, with adaptations for local housing stock, climate conditions, and regulatory frameworks.

Q: What is the payback period for building owners who invest in demand response capability? A: Payback periods vary by building type and scope of investment. For commercial buildings that already have basic building management systems, adding demand response capability through a BEMS upgrade and sub-metering typically costs EUR 15,000 to 40,000 and achieves payback in 2.5 to 4 years through flexibility revenue alone. For residential participants, the payback calculation depends on whether retrofit investments are included. A smart thermostat and heat pump controller installation costing EUR 500 to 1,200 pays back in 3 to 5 years through energy savings and flexibility payments. When combined with insulation and heat pump installation, total investment of EUR 15,000 to 35,000 per household achieves payback in 6 to 9 years at current energy prices, though available subsidies covering 30 to 50% of costs can reduce this to 4 to 6 years.

Sources

• City of Amsterdam. (2025). Flexpower Amsterdam: Five-Year Performance Report and Scaling Strategy. Amsterdam: Gemeente Amsterdam.
• Liander. (2025). Grid Congestion Management Through Demand Flexibility: Network Investment Deferral Analysis 2019-2025. Arnhem: Alliander N.V.
• European Commission. (2024). EU Buildings Performance Report: Energy Consumption, Renovation Rates, and Policy Implementation. Brussels: European Commission.
• CBRE Netherlands. (2025). Commercial Building Flexibility Participation: Tenant Impact Assessment and Revenue Analysis. Amsterdam: CBRE Group.
• ENTSO-E. (2024). Ten-Year Network Development Plan 2024: Scenarios and Grid Investment Needs. Brussels: ENTSO-E.
• Netherlands Enterprise Agency (RVO). (2025). ISDE Subsidy Programme: Residential Energy Retrofit Outcomes 2021-2025. The Hague: RVO.
• Spectral Energy. (2025). AI-Driven Flexibility Dispatch: Performance Metrics from the Flexpower Amsterdam Platform. Amsterdam: Spectral B.V.
• European Commission. (2024). Electricity Market Design Reform: Final Regulation and Implementation Timeline. Brussels: European Commission.