We used to have a problem with too little renewable energy. Getting access to the electrical grid was a secondary concern for project developers. It was first-come-first served, just get into the queue and you would get the connection eventually. As part of the deal, you might have gotten feed-in priority, compensated curtailment or socialized connection costs on top.
We still need more renewable energy, but right now the problem is that we have too little grid.
In Europe alone, more than 1700 GW of renewable and hybrid projects are stuck in connection queues (Energy Storage Europe Association).
In the US, over 2060 GW of generation and storage was actively waiting in interconnection queues at the end of 2025. For projects that actually got built in 2025, the median wait from application to operation was over five years.
German congestion management costs have been running at close to €3 billion a year (€2.9bn in 2024, and €2.2bn over the first three quarters of 2025).
Spain's solar capture factor collapsed to around 18% in February 2026, from 71% in the same month a year earlier (Pexapark).
This means the logic is changing: from first-served to grid-serve. America already made this official: FERC's Order 2023 scrapped first-come-first-served entirely, replacing it with 'first-ready, first-served.' Europe is heading the same direction, just without saying it out loud yet. The projects that get built in the future will be the ones that can prove that they help the grid out in some way, instead of adding to the chaos.
We’ve compiled a list of the five different ways a project can serve the grid:
Asking if there’s capacity at the nearest substation is a naive approach. The grid-serving question is whether your project relieves congestion in the grid. If your project is sited in an area where the grid has a surplus of generation, it might worsen the problem. If you can site your project in an area where there is a lot of load, it might provide crucial relief for the system.
It’s the same technology, but the benefits are very different. Positioning in the grid is becoming as important as irradiance or cost of land for project developers. Germany is going a step further and putting a price on it. The Bundesnetzagentur's grid fee reform is working toward a dynamic tariff, potentially from 2029, where what you pay depends on grid capacity and where your installation sits.
Solar has been so successful that it is destroying value for itself. Everyone produces power in the middle of the day when the sun shines. Deep duck curves are eating away at the capture rates for everyone. Serving the grid means shifting production towards the hours when the power is actually needed. East-west panel orientation might help, but the winner is co-located storage.
Regulation is catching up, with support schemes across Europe moving to strip compensation for production during negative-price hours. Germany's Solarspitzengesetz made that concrete in February 2025: new assets lose their support payment for every hour the wholesale price goes negative, and there were 457 such hours in 2024. Spain shows where that ends up: 41% of April 2026's solar production landed in negative-price hours.
Flexible connection agreements let a developer connect now instead of waiting years for physical grid upgrades. The flexibility is given to the grid operator, and might involve agreeing to caps or curtailments for the developer. Giving up a few hours of production in the middle of the day (when prices are low anyway) in return for building your project now sounds like a great deal.
Brussels is actively pushing this. The Grids Package tells member states how to design flexible connection rules and how to remunerate the developers who accept them, and the storage industry is asking for those agreements to be proportionate, technology-neutral and time-bound rather than open-ended.
Texas has run this experiment for almost 20 years. Under ERCOT's "connect and manage" model, developers pay only their direct connection costs and get connected fast. In exchange, they accept curtailment risk managed through real-time prices. That’s why ERCOT connects projects faster than anywhere else in America. Europe's flexible connection agreements are the same trade with a different accent.
A lot of plants are already connected to the grid. These connections can be reused, with solar, wind and batteries sharing the existing connections. Software does the work of making sure that excess power is not fed into the grid. No new copper in the ground means less to fight over with the grid operator when permitting comes around.
This stopped being theoretical in 2025. Germany legalised cable pooling that January, letting wind and solar operators share a connection and split the capacity-dependent costs. The Dutch regulator got there first and now allows up to four parties to share a connection from 2 MVA upwards, batteries and electrolysers included. For anyone building hybrid or standalone storage, this is the cheapest grid capacity on the market.
Inverters can help the stability of the grid. Reactive power, voltage support and grid-forming capabilities are increasingly a connection requirement. The developers who lead on it get faster, friendlier conversations with operators drowning in applications.
The requirement is close to binding. The amended network code for generators would oblige new storage and renewable modules above 1 MW to be grid-forming, and ENTSO-E published its second technical report on those requirements in November 2025. The urgency traces back to 28 April 2025, when Spain and Portugal went dark for ten hours. ENTSO-E's investigation put the cause down to a loss of voltage control and low system inertia, and explicitly not to renewable generation itself. Every operator in Europe read that report.
None of this is altruism. A grid-serving project connects faster, curtails less and captures more.
Serving the grid is becoming the price of admission. The question is how, and whether you can know before committing to a new project. The data on hosting capacity, local net load, power flows and curtailment risk is fragmented across 44 transmission operators and roughly 2,400 distribution operators beneath them, each with its own formats, licences and refresh cycles (Germany alone is a small research project, and the Nordic grid is another). The US has the opposite disease. The data technically exists, but it's scattered across seven markets and a thousand filings. Public isn't the same as usable.
Developers today typically discover grid value at grid connection application time, when it's too late to change anything. The winners of the next decade will see it at screening time.
This is the problem we are working to solve at Glint Solar.
We're building tools to make grid value visible early, to help you pick the right projects before land is secured and the application is filed. If you want to follow what we're building in this space, and be part of a group of people who care about how our grid actually gets used, subscribe here.