Six Pillars of Consumer Flexibility: What Operating DERs at Scale Across the EU and US Actually Takes

Same customer base. Same device type. One change to the proposition, and a seventeen-fold increase in connected users. That single data point, surfaced by Enode during a recent joint session with Codibly, contradicts most of what utility and retailer programs assume about what scales consumer flexibility. The leverage was not at the device level. It sat in the layers above it: the proposition the customer signs up for, the orchestration architecture that makes the program work in production, and the organisation that has to scale it.

That reframing matters because the conventional answer to “how do we operate distributed energy at scale?” still tends to start at the device layer: hardware coverage, protocol counts, OEM integrations. After an hour with the Enode team, the throughline pointed somewhere else. Device-by-device integration, once the binding constraint, is now substantially commodified — platforms like Enode absorb the long tail of OEM connectivity. The harder engineering work has moved up the stack, into the orchestration architecture that absorbs market fragmentation, coordinates multi-asset dispatch, and carries program logic across mandated and voluntary protocols. That architectural decision sits alongside proposition design and organisational alignment as the three factors that now determine whether a consumer flexibility program scales. For utilities and retailers serious about contributing to the energy transition, that is the work in front of them.

The six pillars of a flexibility program that scales

Capturing the full value of distributed energy resources (DERs) breaks cleanly into six pillars. They are sequenced, but no pillar is optional, and none can be deferred without consequence.

Pillar	What it covers
1. Customer interface	Tariff or program design, native app or portal, customer support at a low cost to serve. The proposition the end-user actually sees and signs up for.
2. Asset connectivity	Cloud-to-cloud or cloud-to-device onboarding, telemetry and control of assets. Where coverage and reliability are non-negotiable.
3. Local optimization	Apply each customer’s constraints (EV departure time, HVAC comfort, self-consumption preferences, tariffs) before the asset is exposed to a market signal.
4. Forecast and aggregation	Predict device behaviour from history and local-optimization configuration, then aggregate into a dispatchable asset pool with statistics disaggregation back to individual assets.
5. Market optimization	Day-ahead, intraday, and imbalance market positioning. Portfolio optimization that builds trader recommendations across markets.
6. Trading	Execute the position. Re-optimize as market conditions and load forecasts change. Usually integrated with the customer’s ETRM or trading system.

The temptation in early-stage programs is to over-invest in the visible pillars (connectivity and trading) while under-investing in the quieter ones. Customer interface gets a thin app. Local optimization defaults to vendor settings. Forecasting becomes a back-office exercise. The program enrolls devices, dispatches some kilowatts, and stalls. The pattern is consistent enough across markets to be predictable. Programs that scale build all six pillars in parallel and treat each as a measurable capability, not a checkbox. (For a deeper architectural view of how distributed energy resource management systems (DERMS), virtual power plants (VPPs), and ADMS relate to each other, see DERMS vs VPP vs ADMS.)

Scaling across 19 million contracts: the case for a unified VPP stack

A useful proof point sits with ENGIE, which Enode powers across the Netherlands, France, and Belgium. The objective there was not to run a flexibility pilot but to operationalize consumer flex as a structural part of energy portfolio management: embedded in trading, generation planning, and customer retention strategy at the same time. The result, per the Enode case study, is a full EV flexibility stack rolled out across more than 19 million residential contracts in the three markets, with thousands of devices migrated in under three months and flex operations now integrated into ENGIE’s existing portfolio management processes.

The number worth dwelling on is not 19 million. It is the migration window. Moving thousands of consumer devices into a multi-market dispatchable VPP in under a quarter is only possible when the connectivity, optimization, and flex layers were architected together from the start. Bolt-on flexibility (connectivity from one vendor, optimization from another, market participation through a third) collapses long before that scale.

Daytime charging at solar peak: the case for multi-device co-optimization

The second public case point is ev.energy, a DERMS partner Enode supports across North America and Europe. In California, where solar penetration is high and midday over-generation is the structural problem, ev.energy has shifted more than 45% of daytime EV charging to peak solar hours across the utility programs it operates. That outcome scales across more than 200,000 end-users and 55+ utilities worldwide. The architecture point underneath is more important than the metric: the program treats EVs, home chargers, solar inverters, and batteries as a single coordinated asset pool, not as four separate device programs.

Single-device flexibility programs leave value on the table. They are easier to launch, simpler to model, and politically easier to fund, yet they fail to capture the co-optimization gains that emerge only when assets are coordinated. That gap matters most in the markets where grid investment deferral is the actual business case for flexibility. A utility that defers infrastructure by orchestrating EVs and batteries together captures multiples of the value of a single-device demand response (DR) program.

Why a fragmented EU and a state-led US need the same kind of integration layer

The most counter-intuitive lesson from cross-market operations is that the EU’s 27-country fragmentation problem and the United States’ state-by-state regulatory mosaic are not different challenges. They are the same architectural challenge expressed in two political contexts.

In California, the choice to mandate IEEE 2030.5 for distributed energy interconnection (through Rule 21 and the CSIP profile) was not a bet on the perfect protocol. It was a bet that one good standard, decided early and enforced consistently, would deliver more value than years of consensus-building toward a theoretically better one. The bet has paid off. Industry analysis of California’s curtailable-inverter deployment under IEEE 2030.5 suggests an additional ~30% of grid headroom has been recovered on the existing infrastructure, capacity that would otherwise have required new build-out (more on the trajectory in IEEE 2030.5 in 2026: From California Mandate to National Trajectory).

Europe’s situation is the inverse. The EU has 27 ancillary services regimes, 27 retail market structures, and Distribution System Operators (DSOs) that each define flexibility procurement on their own terms. Some of that fragmentation is irreducible. Local energy sharing in Portugal works differently than Dutch capacity markets via GOPACS, and that is not going to change. The architectural answer is not to standardise the markets. It is to standardise the integration layer that absorbs the differences, and let the local market logic stay local.

Dimension	United States	European Union
Primary counterparty	Regulated utility (often with retail), DR aggregator, ISO/RTO market.	Deregulated retailer (often the program owner), DSO for local flexibility, TSO for balancing.
Standards baseline	IEEE 2030.5 / CSIP mandated in California (Rule 21) and extending in Texas, Hawaii, Utah; OpenADR 2.0b / 3.0 for DR signalling; CTA-2045 for end-device interoperability.	27 national regimes; balancing platforms MARI / PICASSO standardised at TSO level; DSO flexibility procurement and local energy markets (e.g., GOPACS in NL) defined locally.
Market design pressure	State-by-state mandates, utility-specific program rules, ISO/RTO capacity markets driving compliance investment.	EU-level direction (network codes, AFIR for e-mobility) layered over national retail and grid-fee structures; non-wires alternatives and local energy sharing emerging unevenly.
Integration challenge	Aligning device connectivity with mandated protocols and utility-specific program logic; multi-ISO retailers add complexity.	Absorbing per-country market, retail, and DSO differences behind a unified consumer-facing program layer.
Architectural answer	Standardise on the mandated protocol (IEEE 2030.5, OpenADR), abstract program-specific logic above it.	Standardise the integration layer; keep local market logic local; do not wait for cross-border consensus.

That is the lens Codibly brings to consumer flexibility implementation. The integration layer is the engineering problem with the most leverage in this space right now. It is where market fragmentation, protocol mandates, multi-asset orchestration, and program-specific logic converge, and where most programs that fail to scale have under-invested. In a recent engagement with APG&E, a US retail electricity provider operating across ERCOT, PJM, and NYISO, our team built a custom DERMS platform from the ground up: using Enode for device connectivity, standards-ready for OpenADR and IEEE 2030.5, and designed to extend across additional DER asset classes without re-architecting. The platform reached production in 12 weeks. The point is not the timeline. The point is that the architectural decisions about the integration layer (what abstracts away, what stays exposed, what extends without re-architecting) are what decide whether the platform reaches scale. Device connectivity is the commodity input; the integration architecture is where the program lives or dies. (See also Codibly’s IEEE 2030.5 work with SolarEdge across multiple Australian DNSPs, where the same architectural pattern accelerated market access to fragmented regional standards.)

Where the engineering work has moved

Two patterns sit alongside the architectural work, and both decide whether a well-engineered platform actually delivers in production: proposition design and organisational alignment.

Best-in-class consumer flexibility programs share four operating traits, and most of them sit outside the engineering team. They treat consumer flex as a core strategic pillar of the business, not a side project. They embed flexibility KPIs into how the organisation measures itself: not just enrollment, but participation rate, retention, delivered megawatts, and revenue stacking. They commit cross-functional teams, with shared targets that survive the change of an executive sponsor. And they take a pragmatic posture: progress and iteration beat perfection, and a workable program shipped in six months is worth more than a perfect one stuck in alignment for eighteen.

The opposite pattern is visible in pilot-to-scale failures. A flexibility pilot runs in a standalone application by a small team, hits its enrollment target, and then has to migrate into the retailer’s main mobile app or the utility’s customer platform. At that point it collides with cybersecurity review queues, product roadmap committees, and engineering teams that were never aligned on the scaling KPIs. The pilot was technically successful and organisationally orphaned. The engineering shipped — but the architecture was scoped for a pilot, not for the operating model the business actually needed. That gap, between pilot-grade and production-grade architecture, is exactly where the high-leverage engineering work now lives.

There is a related discipline on the customer side: participation curves. Con Edison’s published demand response data shows what happens when programs treat enrollment as the success metric. First-event participation is near-universal. Over subsequent events, opt-outs accumulate on an exponential curve. By the time the utility needs the capacity at scale, the participation base has eroded, sometimes by half or more, because the program was not designed for sustained engagement. The fix is not better dispatch logic. It is layered, segmented program design that gives customers multiple ways and reasons to stay in. (Codibly works with utilities and aggregators on exactly this kind of program-design and platform integration through our Demand Response Programs and Optimizers & Aggregators practice.)

Where this lands: standardise the base layer, scale the business on top

A useful way to think about the next phase of consumer flexibility is the analogy that closed the joint session. The internet works because the base communication layer (TCP/IP) was standardised early, and all the business and product innovation that has happened since happened on top of that layer, not in competition with it. Nobody is trying to redesign basic packet routing in 2026. The trillion-dollar economy that runs on the internet exists because that argument was settled.

Distributed energy is approaching the same inflection point. The base layer is converging: interoperable device connectivity through protocols like IEEE 2030.5, OpenADR 3.0, and Open Charge Point Protocol (OCPP), supported by integration platforms that absorb market fragmentation. That convergence is engineering work — protocol implementations, certification programs, accelerator codebases, and integration architectures that have to be specified, built, and operated by someone. What sits on top of the base layer is where business innovation will happen over the next decade: propositions that move retention curves, multi-device co-optimization that captures grid-investment deferral value, and market designs that compensate flexibility properly. The role of platforms, retailers, utilities, and integration partners is to commit to the base-layer engineering hard enough that everyone can stop arguing about it, then compete on what actually matters to customers and the grid.

The next twelve months will likely settle, in many markets, which integration layer wins by default. California has already shown what early regulatory clarity buys. The EU’s 27-country fragmentation will not resolve through consensus alone, but it can be absorbed by an architecture choice. That architecture is the work in front of every utility, retailer, and aggregator with serious flexibility ambitions, and the conversation worth having now.