For most of the past decade, the defining question for an energy project was the cost of a megawatt. Today the harder question is whether that megawatt can reach anyone, reliably, when it is needed. Electricity demand is rising, renewable generation is being built at record pace, and data centres and electrified industry want large connections quickly. The networks, equipment and approvals that tie all of this together have not kept up.
The central argument of this article is that the critical path of most energy infrastructure projects now runs through the grid rather than the generating asset: through connection queues, transformer and cable factories, permitting and system operation. By the International Energy Agency’s latest count, at least 1,700 GW of advanced-stage renewable projects and 600 GW of utility-scale batteries were awaiting grid connection in 2025 (IEA, Modernising Grids in the Age of Electricity, 2026). The programmes that deliver, and keep delivering when something fails, plan generation, networks, storage and demand as one portfolio, secure scarce equipment on programme horizons, extract more from existing networks while new lines are built, and design for the ways power systems break.
The measure of success changes accordingly. Nameplate megawatts installed matter less than reliable, connected capacity and the resilience of the system it joins.
In brief
- The bottleneck has moved from generation to the grid. In the United States, the median project completed in 2025 took 61 months from interconnection request to commercial operation, against 22 months in 2008 (Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition).
- Equipment is the new critical path. Large power transformers can take up to four years to secure and cables two to three, with lead times almost doubling since 2021 (IEA, Building the Future Transmission Grid, 2025).
- Grid spending is rising, partly because equipment costs more. Global grid investment reached nearly US$450 billion in 2025, but some of the increase reflects higher equipment prices rather than more infrastructure (IEA, World Energy Investment 2026).
- Resilience is designed and operated in. The 2025 blackouts in the Iberian Peninsula and Chile, and the substation fire that closed Heathrow Airport, show how single faults cascade through dependent systems.
- The quickest capacity often comes from the network already in place. The IEA estimates that better use of existing grids could connect up to 330 GW of new generation, storage and demand without reinforcement.
Why the grid is now the pacing item
The IEA forecasts global electricity demand to grow by an average of 3.6% a year from 2026 to 2030, driven by industry, electric vehicles, air conditioning and data centres, which would make annual growth over the next five years about 50% higher than the average of the previous decade (IEA, Electricity 2026).
The International Renewable Energy Agency reports that the world added 692 GW of renewable power capacity in 2025, a 15.5% increase, taking the total to 5,149 GW; solar alone accounted for 511 GW (IRENA, Renewable Capacity Highlights, 2026).
The difficulty is timing. The IEA found that new grid infrastructure often takes five to 15 years to plan, permit and complete, compared with one to five years for new renewable projects, and that a single extra-high-voltage line in an advanced economy can take five to 13 years to pass through permitting and construction (IEA, Electricity Grids and Secure Energy Transitions, 2023). In practice, generation that can be built in two years may wait on a network that takes ten. Under current policies, the IEA estimates that meeting demand growth to 2035 requires grid capacity to rise by at least 30%, equivalent to adding or replacing 25 million kilometres of lines (IEA, Modernising Grids in the Age of Electricity, 2026).
Investment is climbing, but less than the headline suggests. The IEA’s World Energy Investment 2026 reports that global grid spending rose by US$45 billion in 2025 to nearly US$450 billion and is projected to grow by a further 17% in 2026. It cautions that the upswing partly reflects inflation in grid equipment rather than more infrastructure delivered, and that spending remains below requirements. Its companion Electricity 2026 report estimates that annual grid investment must rise by roughly 50% by 2030 to meet forecast demand. The wider pipeline of energy and infrastructure projects behind these numbers is set out in our analysis of the next decade of global projects.
The connection queue is a portfolio problem
The mismatch is starkly visible in the United States. Lawrence Berkeley National Laboratory’s Queued Up: 2026 Edition found roughly 2,061 GW of capacity actively seeking interconnection at the end of 2025: 1,312 GW of generation and 749 GW of storage. Only 13% of the capacity that requested interconnection between 2000 and 2020 had reached commercial operation by the end of 2025, while 75% had been withdrawn. Of the 2,290 GW active in queues at the end of 2024, just 53 GW became operational in 2025, a throughput of about 2%.
Our reading is that a connection queue is a portfolio without portfolio management. Under first-come, first-served rules, a speculative application occupies the same place as a shovel-ready project, network studies are clogged with schemes that will never be built, and viable projects wait behind them. The answer is the same discipline any serious investment portfolio needs: entry criteria, prioritisation against strategic need, and the willingness to stop what should not proceed, the subject of our Insight on what portfolios should stop.
Great Britain has attempted exactly that. Its National Energy System Operator (NESO) reported in December 2025 that the old first-come, first-served model had let the connections queue grow tenfold in five years, to more than 700 GW of generation and storage, roughly four times what the country needs for its 2030 power system. After assessing around 3,000 applications against readiness and strategic alignment, NESO confirmed a new pipeline of 283 GW of generation and storage and almost 100 GW of transmission-connected demand; 132 GW was identified as aligned with the government’s Clean Power 2030 target, and more than 300 GW from the old queue will not move forward for now. NESO was clear that connection reform is no substitute for building thousands of kilometres of new lines and cables, or for faster planning decisions.
The queue problem is no longer confined to generators. The IEA reports that large-load connection queues are lengthening rapidly at US grid operators including ERCOT and PJM, largely because of data centres, and that the Netherlands and the United Kingdom face delays connecting industrial consumers and electric vehicle charging (IEA, World Energy Investment 2026). For portfolio leaders, grid access is now a portfolio constraint in its own right.
Equipment and supply chains: the new critical path
When projects do clear the queue, they meet a second bottleneck. An IEA survey of industry for Building the Future Transmission Grid (2025) found that it now takes two to three years to procure cables and up to four years to secure large power transformers, with average lead times almost doubling since 2021. Direct current cables, often preferred for long-distance links, can take more than five years. The same survey suggests cable prices have nearly doubled since 2019 and power transformer prices have risen by around 75%. Permitting remains the main cause of delay, particularly in advanced economies, but component supply is increasingly a limiting factor.
Manufacturers are adding capacity, with capital expenditure in the first three quarters of 2025 some 12% higher than a year earlier, and utilities are shifting to framework agreements that secure supply over longer horizons instead of contracting project by project (IEA, World Energy Investment 2026). The IEA’s transmission report makes the connection explicit: manufacturers will invest when national and regional transmission plans are visible and credible enough to translate into firm component demand. It also estimates that the roughly 8 million people who build, maintain and operate grids worldwide will need to grow by 1.5 million by 2030.
For capital projects leaders, the implication is that procurement strategy moves upstream and becomes a programme decision rather than a project task. Standardising specifications across a portfolio, committing to framework volumes, reserving factory slots before final consents and freezing designs earlier all trade some flexibility for schedule certainty. Long-lead equipment, connection dates and consents belong on the critical path of every integrated schedule, which is the everyday work of project controls.
Resilience: designing for the day things go wrong
Energy infrastructure is judged hardest when it fails. The IEA’s Electricity 2026 catalogues a year of instructive outages. On 28 April 2025 the Iberian Peninsula suffered the largest European blackout since 2003; according to the initial investigations cited by the IEA, it had multiple causes, including high voltage volatility, limited reactive power absorption, power oscillations and rapid disconnections, and the system collapsed within seconds. Interconnections with France and Morocco, alongside domestic black-start units, helped restore supply in Spain within 12 to 16 hours. On 25 February 2025 a protection system failure shut Chile’s main north–south transmission line, leaving 14 of its 16 regions, home to about 98% of the population, without power, at an estimated cost of US$450 million. On 21 March 2025 a fire in a single transformer at a substation near Heathrow closed the airport for a day, affecting more than 300,000 passengers and 1,350 flights, even though the airport had three supply routes.
Three lessons follow for project design. First, stability is becoming as important as capacity: the IEA highlights voltage management as increasingly important for power system stability, so system services belong in project specifications, not only megawatts. Second, redundancy on paper is not resilience in practice; the Heathrow case shows that several supply routes do not guarantee continuity if the loss of one critical asset cannot be absorbed. Third, interconnection is a resilience asset as well as a trading one.
Age compounds the risk. The IEA found that more than half of grid infrastructure in advanced economies is over 20 years old, and that transformers typically have design lives of 30 to 40 years; proper maintenance can extend equipment life significantly, while deferred maintenance leads to premature failure (IEA, Electricity Grids and Secure Energy Transitions, 2023). Utilities are responding: the IEA estimates that European utilities now direct 25–40% of their annual capital expenditure to grid resilience and modernisation (IEA, World Energy Investment 2026).
Maintenance choices can matter more than structural upgrades. The World Bank’s Lifelines (2019) notes that during storms, flying debris and vegetation, rather than wind itself, are the main causes of damage to power poles, so trimming trees can be more efficient than reinforcing poles. When Hurricanes Irma and Maria struck Puerto Rico in 2017, trees falling on transmission lines were a major reason the grid was so badly damaged, and every customer of the island’s power authority lost supply for more than a week. The wider economic consequences of unreliable infrastructure are explored in global projects, real-world impact.
Security of supply now carries a geopolitical premium as well. The IEA’s World Energy Investment 2026, published amid conflict in the Middle East that has shaken confidence in transit through the Strait of Hormuz, notes that alternative routes, stronger supply chains, redundancy and inventories all make energy systems more secure but also more expensive, and that the cost-efficiency of security measures will become central to energy debates. Resilience is worth paying for; the discipline is in buying the right kind.
Getting more from the network you already have
The fastest megawatt is often the one connected to a line that already exists. The IEA’s Modernising Grids in the Age of Electricity (2026) estimates that dynamic line and transformer ratings, topology optimisation and advanced power-flow control could allow up to 330 GW of additional generation, storage and demand to connect to existing networks without reinforcement. Connecting the same volume through network expansion would require around US$100 billion of investment. Congestion is already costly: US$12 billion in the United States and €4.3 billion in the European Union in 2024.
The report also notes that removing all congestion would mean building a grid sized for peak conditions that sits underused most of the year, so a degree of congestion is efficient. Separately, the IEA’s Electricity 2026 estimates that more flexible, non-firm connection agreements alone could enable 750 to 900 GW of advanced-stage projects to connect sooner, accepting some limits on output until reinforcements arrive.
Technology is not the main barrier: in an IEA survey of 25 network operators, 64% cited skills and organisational readiness as a major obstacle to deploying these tools, ahead of data quality and trust. That makes grid modernisation as much a transformation programme as an engineering one.
Industrial energy infrastructure: when the customer becomes the constraint
Industrial and digital companies are now energy infrastructure developers whether they intend to be or not. The IEA reports that technology companies accounted for around 40% of corporate power purchase agreements signed worldwide in 2025, and that data-centre build-outs now carry their own spending on power equipment, grid-connected and on-site generation and grid upgrades. Technology sector demand also lay behind nearly all of the US$28 billion of gas turbines ordered for on-site generation as part of the US rebound in gas-fired investment (IEA, World Energy Investment 2026).
For any industrial capital project, from a data centre to an electrified process plant, the grid connection has become a gating item for the final investment decision. Leaders should treat power as long-lead infrastructure: screen sites for available capacity and realistic connection dates before committing, consider flexible connections or on-site generation and storage as a bridge, and build the connection timeline into the business case. These are central questions for capital projects and CAPEX leaders.
A delivery model for resilient energy projects
| Stage | Conventional practice | Resilient practice |
|---|---|---|
| Site and connection | Choose the best resource, then apply to connect | Screen sites for grid capacity and connection date first |
| Queue position | Hold speculative positions open | Enter with evidence of readiness; release positions that will not proceed |
| Equipment | Tender project by project after consent | Framework agreements, standard specifications, reserved factory slots |
| Design | Optimise for nameplate output | Specify system services such as voltage support alongside output |
| Controls | Track construction progress | Track connection, consents and long-lead equipment as the critical path |
| Operations | Repair after failure | Condition monitoring, planned maintenance and rehearsed restoration |
| Value measure | Megawatts installed | Reliable, connected capacity and contribution to system resilience |
What boards and energy leaders should do
- Utility and network executives should publish credible, long-horizon investment plans that suppliers can build factories against, pair expansion with grid-enhancing technologies and flexible connections, and fund maintenance of ageing transformers and lines as a resilience investment.
- Developers and independent power producers should manage queue positions as a portfolio, withdrawing early from weak sites, and bring connection, consent and equipment risk into the earliest stage of project appraisal.
- Industrial and data centre owners should make power availability a site-selection criterion rather than a later workstream, and plan bridging solutions before committing capital.
- CFOs and investors should value energy projects on connected, deliverable capacity and scrutinise schedules for unsecured long-lead equipment.
- Governments and regulators should reward readiness over queue position, allow anticipatory grid investment, and set resilience standards that treat interdependent systems, such as airports and hospitals, as part of the power system.
Across the project economy
Energy infrastructure concentrates the challenges of every area of the project economy. Decide means choosing which projects deserve scarce connection capacity. Fund means financing grids ahead of need and paying a justified premium for resilience. Deliver means managing equipment, permits and connection dates as one critical path. Prove means measuring reliable, connected capacity and performance under stress rather than megawatts commissioned. The broader principles of pipelines, maintenance and delivery bodies are set out in our companion piece on infrastructure investment; energy is where they are tested most severely, because a failure in power can spread almost immediately into transport, water, health and digital services, as Heathrow showed.
Conclusion: connected time is the scarce resource
Capital and generation technology still matter, but in many markets they are no longer the scarcest inputs. What is scarce is connected time: the years between a decision to build and a secure, working connection to a system that can use the output. Queue reform, equipment frameworks and smarter use of existing lines are all ways of buying back that time. The organisations that treat connected time as an asset, managed across a portfolio rather than fought for project by project, will deliver more reliable power sooner, and will be the ones still delivering when the system is under strain.
Frequently asked questions
What are energy infrastructure projects?
Energy infrastructure projects build or renew the systems that produce, move and store energy: power stations, renewable plants, transmission and distribution grids, interconnectors, substations and batteries, and increasingly the connections and on-site power of large industrial and digital users.
Why are grid connection queues so long?
Generation and storage can be built far faster than the grid they connect to, and first-come, first-served rules let speculative applications occupy capacity alongside viable ones. Transformer and cable shortages, permitting and study backlogs add further delay, which is why reforms now prioritise projects on readiness and strategic need.
What makes power infrastructure resilient?
Resilience comes from avoiding single points of failure, managing voltage and stability as well as capacity, maintaining ageing assets, interconnecting with neighbouring systems and rehearsing restoration. It is designed in from the start and sustained through operation.
Sources
- International Energy Agency, Building the Future Transmission Grid (2025). IEA
- International Energy Agency, Electricity 2026 (2026). IEA
- International Energy Agency, Electricity Grids and Secure Energy Transitions (2023). IEA
- International Energy Agency, Modernising Grids in the Age of Electricity (2026). IEA
- International Energy Agency, World Energy Investment 2026 (2026). IEA
- International Renewable Energy Agency, Renewable Capacity Highlights (2026). IRENA
- Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition, Characteristics of Power Plants Seeking Transmission Interconnection as of the End of 2025 (2026). Berkeley Lab
- National Energy System Operator, NESO implements electricity grid connection reforms to unlock investment in Great Britain (2025). NESO
- World Bank, Lifelines: The Resilient Infrastructure Opportunity, by Hallegatte, Rentschler and Rozenberg (2019). World Bank