Why this recommendation matters
Firm connection rights are an inefficient, costly default. Offering the right to a firm connection reserves grid capacity for the maximum load a project could ever import or export. When evaluating whether such a connection can be granted, system operators then typically study whether that capacity is available in the worst conditions the grid would face, even though these conditions tend to occur for only a few hours a year, if at all. This usually results in a generous safety margin; that is to say a significant share of grid bandwidth is reserved for events that rarely, if ever, materialise. This approach made sense in an era of limited grid visibility and remote control. As connection timelines now run to years and grid reinforcement to a decade, it becomes imperative to use this bandwidth more efficiently. This is capacity the system already has and could allocate given the right interventions. One part of the equation is growing the system operator’s capabilities, as addressed under objective 5. The other part is leveraging the flexibility on the side of the party requesting the connection.
This recommendation calls for a reciprocal arrangement in place of the firm default. Where a connecting party is willing to accept operational constraints, the grid company can offer terms that would otherwise be unavailable. Flexible connection agreements grant access in exchange for accepting occasional curtailment or time-varying limits. Co-located arrangements let complementary supply, storage and demand share a single connection point, with combined export managed within its capacity.
Phased agreements release capacity in step with demand that materialises over years rather than all at once. Locational signals and transparent connection-cost allocation make the reward legible, so that a developer choosing a site where headroom exists sees the benefit in time and cost. Dynamic operating envelopes are the mature form of the same idea – they adjust import and export limits in real time according to network conditions, as opposed to setting fixed limits that align with worst-case assumptions.
The gains are immediate and shared. Projects connect sooner when grid companies reward flexibility, which keeps electrification and renewable deployment on schedule and reduces the financing cost of waiting. Reinforcement is deferred or avoided where flexibility does the same job, and the avoided expenditure is outlay consumers would otherwise fund through grid charges. Grid operators gain additional means of managing constraints, since curtailable, co-located and dynamically managed connections give them levers that a fleet of firm connections does not. The flexible connection terms agreed at the point of connection add to the flexibility as mandated by the standards and grid codes in recommendation 5.1, and to wider activation and coordination of the provision of flexibility and other system services discussed under objective 6.
How to act on this recommendation
Stage 1
Make flexible access a standard option rather than an exception.
Regulators should introduce flexible, non-firm connection agreements as an early and routinely available choice, granting faster access in exchange for accepting curtailment or time-varying limits. They should also enable co-located connection arrangements that let complementary assets share a connection point. Offer these arrangements as much as possible through standardised, published template terms, rather than negotiating them case by case, to avoid introducing a new source of administrative complexity and fragmentation. At the same time set a transparent, regulated methodology for allocating connection costs, since it determines what a connecting party pays and is the basis on which any reward for flexibility is calculated. Large loads warrant their own treatment, with connection terms that offer faster access in exchange for more binding flexibility commitments such as curtailment during scarcity, on-site generation or storage, or contracted demand response. The technical requirements those loads must meet are set under recommendation 4.1, and the process by which they are assessed is in recommendation 4.2.
Stage 2
Extend arrangements to cover capacity that arrives over time and the location where projects choose to site.
Frameworks for phased connection agreements should release capacity in line with planned demand growth, so that a project can secure a connection path without reserving its full capacity from the outset. Projects that connect where capacity is available should be rewarded with faster and/or cheaper access, using locational signals in connection charges together with the spatial planning that steers new connections towards the parts of the grid with headroom. Both draw on the planning methodologies in objective 1 and the spatial designations in recommendations 2.1 and 2.2. They also depend on the hosting capacity information published under recommendation 4.2.
Stage 3
Make access dynamic.
Dynamic operating envelopes set connection limits that vary with actual grid conditions, calculated close to real time. This approach captures headroom that static flexible agreements cannot reach. They depend on the real-time visibility, state estimation and operational capability built under objective 5, on the dynamic hosting capacity developed under recommendation 4.2, and on the capability requirements set under recommendation 4.1, since an asset can only follow a dynamic limit if it was built to receive and act on one.
Core measures
Implementation considerations
Differentiated terms create winners and losers across project technologies, sizes and locations, requiring the reforms to have a transparent, defensible basis. Low-impact and flexible projects gain faster access and simpler procedures, while those insisting on firm capacity may face longer timelines or higher costs, which raises fairness and investment-certainty concerns among large industrial consumers and developers holding existing queue positions. The reforms therefore require a transparent methodology for assessing grid impact and for calculating what flexibility is worth. Unclear definitions or thresholds invite disputes, inconsistent application across regions and reduced investor confidence. The choice of cost-allocation method, in particular how much of a triggered upgrade the connecting party bears, shapes both who pays and where projects locate, and should be settled openly as a policy decision.
Safeguards are needed to prevent flexible connection instruments from concentrating access to capacity. Well-resourced actors are better placed to navigate flexibility and co-location arrangements. Standard published terms rather than bilateral negotiation help smaller developers and community projects use the same options, and they keep a rising volume of flexible connections from fragmenting into inconsistent, hard-to-compare deals. Stakeholder preferences also diverge: developers of storage, EV charging and some industrial loads tend to welcome earlier access even with operational limits, while large generators and constant-load customers value firm capacity and may see flexibility constraints as reducing project value. There are also broader system efficiency trade-offs to consider. Co-location models and phased connections can improve local grid utilisation, but may shift constraints elsewhere if they are not integrated into wider planning frameworks. Cable pooling typically enjoys broad support because it increases the utilisation of existing infrastructure, but it can create coordination challenges related to operating rights and curtailment arrangements among co-located assets. Similarly, hosting capacity maps can improve transparency and decision-making, but grid companies may have concerns about data quality, cybersecurity and the risk of misinterpretation if information is outdated or oversimplified.