Static vs Dynamic Charging Tariff, Which Performs Better

Two charge points with the same price per kWh. At one, drivers wait on Friday evening. The other sits idle for most of the afternoon. A static tariff treats both situations the same, because it is one number for every hour of the week. Whether a dynamic tariff does better has an honest answer: on most public networks it does, but there are sites where static wins, and you want to know which kind you run.

Which performs better, a static or a dynamic charging tariff?

A demand-aware dynamic tariff performs better wherever occupancy clearly varies by hour or by day, because one fixed price is too low when drivers queue and too high when the charger stands empty. A static tariff performs just as well on sites with flat, low demand, on sites with a small group of returning users, and on networks where tariff changes do not reliably reach roaming partners. The strongest setup combines the two ideas: a base tariff that moves by a published rule, within a floor and a ceiling, and that is fixed the moment a session starts. A tariff that only tracks the energy market does not meet that description and belongs on a member rate.

A static charging tariff is a single price per kWh, sometimes with an occupancy fee, that applies at every hour until the operator changes it by hand. A dynamic charging tariff is a price that changes by hour, day or occupancy according to a rule set in advance. Both can be transparent. The difference is whether the price can respond to demand at all.

What does one fixed price do on two different streets?

It overprices one location and underprices the other. The clearest public evidence does not come from charging but from parking, where San Francisco ran the largest test of demand-based pricing for on-street spaces. In July 2011, Chestnut Street and Lombard Street, two parallel streets, both charged $2 an hour. All five blocks on Chestnut were above 80 percent occupancy. On Lombard, two of the five blocks were below 60 percent.

The SFpark programme then adjusted prices in small steps toward a target occupancy of 60 to 80 percent. According to the evaluation by Gregory Pierce and Donald Shoup, published in 2013, the average price on Chestnut had risen 75 percent to $3.50 an hour after 10 price changes in two years, while Lombard had fallen 50 percent to $1.00. Over the first two years, prices went up in 31 percent of the cases, down in 30 percent and stayed the same in 39 percent. The average price fell 4 percent, and by August 2013, 62 percent of blocks were in the target range.

A kerbside bay and a charge point share the property that matters here: capacity that expires by the hour. An empty hour cannot be sold later. The same study found an average price elasticity of minus 0.4 across 5,294 measurements, with large differences by block and by time of day. Price elasticity is how strongly demand responds to a change in price. That spread is the real argument against a single static price. No one number is right for every hour and every location.

Where does a static tariff win?

A static tariff wins where there is little demand variation to steer, or where moving the price costs more than it earns. There are four situations in which we would advise an operator to keep it simple, at least for now.

• Low, flat occupancy. If a charger has few sessions and no visible peak, a dynamic model has almost nothing to learn from and nothing to shift. The work then sits in utilisation first, as we explained in our piece on charge point utilisation and break-even.
• A closed group of returning users. At a workplace or a residential hub, predictability is part of what the user pays for. A monthly subscription often does more there than a price that moves.
• Charging as a side service. At a hotel or a restaurant, the stay sets the dwell time, not the price. See our analysis of EV charging profitability at hotels.
• No control over tariff propagation. A price that moves but arrives wrong at an eMSP costs more trust than a static price ever will. Sent and applied are two different events.

Predictability also has a value that customers will pay for. Anja Lambrecht and Bernd Skiera studied 10,882 customers of a European internet provider and published their findings on tariff-choice biases in the Journal of Marketing Research in 2006. More than half of the customers with a flat-rate bias paid at least 100 percent more than they would have on the cheapest tariff for their usage, and that bias did not significantly increase churn. Internet access is not charging. The lesson still carries over: a dynamic tariff that feels unpredictable loses something a static tariff gives away for free.

How do static and dynamic tariffs compare on a public site?

On a site with varying demand, dynamic wins on revenue and availability, and static wins on simplicity. The table sets out where each one stands.

What goes wrong with dynamic tariffs, and what fixes it?

Most failures come from the design, not from the idea of a moving price. Drivers accept variation they can see coming. In the Dutch national charging survey of 2025, run by RVO, ElaadNL and the Dutch EV drivers association among more than 4,500 drivers, nearly 80 percent said it is important to know the tariff of a public charge point in advance. At the same time, 28 percent usually did not know what they pay for public charging near their home. The same survey found that at public charge points, drivers prefer smart tariffs over a smart, variable charging speed.

• The price changes during a session. Fix the price when the cable goes in and hold it until the session ends.
• Nobody can predict the price. Publish the rule, for example an off-peak block from 22:00 to 06:00, and not only the current number.
• The price climbs without limit. Set a floor and a ceiling. The upper bound is what separates dynamic pricing from surge pricing, as covered in how to set a price floor and ceiling.
• Changes arrive late or wrong at partners. Let changes take effect at a day boundary and verify afterwards that each partner applied them, which is what a nightly tariff deviation check is for.
• The price follows the wrong signal. Base the tariff on demand at your own site, and treat energy cost as an input for the floor.

Where does a tariff that tracks the energy price fit?

As a discount layer for members and subscribers, and not as your public price. A spot price with a fixed margin on top exposes your whole margin to the swings of the electricity market. It also says nothing about whether your chargers are full or empty. For a returning member who reads the rule once and plans around it, that same mechanism works well, and it rewards the drivers who bring you predictable volume.

The base and roaming tariff should be demand-aware instead. It moves with hour, day and occupancy, stays within limits you set yourself, and is at every moment one number the driver sees before plugging in. That is the configuration in which a dynamic tariff reliably outperforms a static one. How that model is built is described on the page about how the Proxilink platform sets prices.

What would you do differently on Monday?

Pull the occupancy per hour for each charger over the last three months and look at the spread. If the hours look alike and utilisation is low, keep the static tariff, put the effort into visibility and uptime, and look again in six months. If you see clear peaks and clear quiet hours, start with a simple time-of-day schedule within a floor and a ceiling, before moving to prices that follow live occupancy. Whichever you choose, check that the tariff you set is the tariff your partners bill. Do you know today which of your sites belongs in which group? The free potential analysis works it out on your own figures.