Grid capacity has become the hard ceiling on public charging. Industry analysis this year puts it plainly: securing grid capacity is now the most critical bottleneck as operators scale high-power networks, and integrating battery storage at charging sites has become crucial to deploy more high-power chargers inside the capacity a site already has. Operators looking at a long wait for a heavier connection are not debating whether to buffer. They are debating how big.
That decision gets made as an engineering decision. It is also, quietly, a pricing decision, and the pricing half usually goes unmade.
A battery breaks the one thing a flat tariff assumes
A flat retail tariff is a bet that your cost per kWh is roughly the same all the time. On a plain grid connection that bet is already shaky, because your import price and your peak-power exposure both move. Put a buffer between the grid and the connectors and it stops being true in a more structural way.
With storage on site you have two different products flowing out of the same cable. Sometimes you are delivering energy the battery absorbed during a cheap, low-stress window, at a marginal cost you locked in hours earlier. Sometimes the buffer is drained and you are delivering energy pulled straight off the meter, at whatever the grid costs right now, at exactly the moment your site is closest to its capacity limit and your peak-power charge is most at risk.
Those are not the same kWh. A single number in your tariff field averages them together and prices both wrong. It sits too high on the hours where you are serving cheaply from the buffer, turning away sessions you would happily have taken. It sits too low on the hours where you are exposed, inviting exactly the demand you can least afford to serve.
The battery's payback depends on a signal you are not sending
The business case for on-site storage is arbitrage plus capacity relief. It earns by absorbing energy when energy is cheap and the site is quiet, and releasing it when energy is expensive or the connection is tight. That case is written assuming the battery gets a reasonable chance to refill between the busy windows.
Retail pricing is one of the few levers you have over when demand actually shows up. If your price says nothing about buffer availability, drivers have no reason not to arrive during the window that drains it fastest, and you refill at the worst possible price to be ready for the next one. The asset still works, but it works against a demand curve nobody shaped. You paid for the flexibility and then removed the mechanism that would have used it.
This is the same plumbing gap that shows up everywhere else in charging economics. Reading a live signal, turning it into a decision at the connector, doing it every few minutes without a person in the loop. On a buffered site there is simply one more signal in the mix, the state of charge of your own battery, and it happens to be the one signal you fully control and fully observe.
We wrote earlier about flexible grid connections, where the network operator is allowed to curtail you at your busiest moment. A buffer is the operator's answer to that same ceiling. But a buffer your pricing does not know about only softens the ceiling physically, not economically.
What this looks like in practice
Three things, in order, and none of them require ripping anything out.
Know your marginal cost by hour, not by month. If the only cost figure in your pricing conversation is a blended annual number, a battery cannot show up in your pricing at all, because its entire effect is to move cost between hours.
Make availability visible in the price, even coarsely. A two level structure that separates the hours your buffer reliably covers from the hours it does not is not elegant, but it captures most of the value of the idea and can be reasoned about by a human. Refinement comes later.
Keep the base tariff demand-aware, not purely energy-linked. It is tempting on a battery site to wire the retail price straight to the spot price, because the arbitrage story makes that feel natural. We would push back on that as your standard rate. A tariff driven only by the energy market exposes your margin fully to market volatility and says nothing about whether anyone actually wants to charge at that hour. Energy-linked pricing is genuinely attractive as a member or subscriber discount layer sitting on top of a demand-aware base rate, where it becomes a benefit you can explain in one sentence. It is a poor choice for the rate every ad-hoc driver pays by default.
If the battery is still on paper
The best moment to have this conversation is before the storage is sized, not after. Sizing assumes a demand profile. Pricing shapes that demand profile. Doing the two independently means you size for the shape you happen to get instead of the shape you could have asked for, and on a system quoted per kWh of capacity that gap is expensive in a way that never shows up as a line item.
The same insight sits next to the point we made about grid flexibility as a second revenue line: it is the same pricing layer opening the door each time. Installing a battery without that layer means buying the hardware of the answer and leaving the answer itself behind.
Sources: Europe EV Charging Report 2026, gridX and What It Takes for CPOs to Scale EV Charging Networks in 2026, Driivz.
Frequently asked questions
Why are batteries appearing on charging sites?
Because grid capacity has become the hard constraint on new fast charging sites. A heavier connection can take a long time in many places, while a buffer between the connection and the chargers allows high power on a modest connection. Industry analysis this year calls securing grid capacity the most critical bottleneck for operators scaling high-power networks, and describes on-site battery storage as having become crucial to deploy more fast chargers within existing limits.
What does a buffer change about your cost structure?
From that point on you are selling two different things through the same cable. Sometimes you deliver energy the battery absorbed earlier during a cheap, quiet window, at a cost locked in hours ago. Sometimes the buffer is empty and you pull straight off the meter, at whatever the grid costs right now, exactly when the site is closest to its capacity limit. Those are not comparable kWh.
What does a flat tariff get wrong about that?
A single number averages both states and prices both wrong. It sits too high on the hours where you serve cheaply from the buffer, so you turn away sessions you would gladly have taken. It sits too low on the hours where you are exposed, so you attract exactly the demand you can least afford to serve.
So should the charging rate follow the spot price?
Not as your base rate. A tariff driven only by the energy market exposes your full margin to market volatility and says nothing about whether anyone wants to charge that hour. Cheap energy at three in the morning does not create demand at three in the morning. Energy-linked pricing does work well as a discount layer for members or subscribers, sitting on top of a demand-aware base rate.