Interview: Discussing the initial success of National Grid ESO’s Demand Flexibility Service

This interview was originally published by Current± and can be read in full here.

National Grid’s Demand Flexibility Service (DFS) has widely been touted as a success by reducing energy consumption around peak demand periods.

The last DFS trial period was scheduled for the evening of 23 March in what was the 19th test event since the scheme began in November 2022.

The DFS service saw a Guaranteed Acceptance Price (GAP) of £3,000/MWh (£3/kWh) for test events, giving consumers a strong price signal to reduce demand against their average historic consumption for the event period.

As well as the test events, there have also been two live uses of DFS. The first ran from 17:00 and 18:00 on 23 January, and saw the operator request 323MW in the first half an hour and 336MW in the second. The second live event ran just the day after on 24 January between 16:30 and 18:00, with 274MW required for the first half an hour, 330MW for the second and 341MW for the final half an hour. Combined the live events have delivered c.0.6GWh of demand reduction.

With the DFS scheme being regarded as a success by a number of energy suppliers including Octopus Energy, Current± spoke with Mark Hamilton, managing director of FlexiGrid at smart energy solutions company SMS plc, who were the largest independent aggregator to participate in the service.

SMS partnered with four different B2C businesses to bring nearly 50,000 customers into the DFS service, demonstrating the important role aggregators will play in unlocking the value of domestic and commercial consumer flexibility in the future.

Could you provide an overview of the DFS trial service and how it has been received by the industry?

I think we’re turning a corner in terms of demand-side flexibility. Largely because of the push from DFS, we’ve suddenly got a situation where consumers and lots of others in the industry seem to now understand what flexibility is.

I think it’s a real stepping-stone in terms of automated flexibility asset participation in particular, where rather than changing their behaviour, consumers can instead participate in grid balancing simply by making their flexibility assets such as batteries, EV charging and electric heating, available to aggregation programmes. We had about 50,000 customers that we were managing in DFS in terms of smart meter data collection to allow them to participate. The majority of these were manual turn-down customers, but we did have some automated participation with domestic battery customers through our partner SolarEdge.

We brought their customers into DFS and for those customers, participation was very simple. Battery customers didn’t need to take any manual action to reduce demand during a DFS event, they just handed over the flexibility of their battery to us as the aggregator, and the battery did all the work for them. We set a schedule on the battery to maximise the value of the service by limiting grid import and in some cases maximising grid export during the DFS period, while the customer consumed electricity as normal.

Battery customers were typically generating 1,000% higher demand reduction during DFS events compared to manual actions, meaning 10 times more revenue versus manual turn-down customers. It’s small volumes for now in terms of automated asset participation in DFS, but it really demonstrates that in the future, the vast majority of demand flexibility is going to be through automated asset control. Consumer driven behavioural change is interesting and can play a part, but automated control is where the real opportunity lies.

In terms of automated asset control for service like DFS, are there some cases where the required functionality is already in place on the asset? Or is it something that will have to be retrofitted into the software? How easy is it to adapt the systems that are in place already to maximise the value from services like DFS?

If a customer knows what they’re doing, they can certainly adapt the scheduling of their batteries, EV charging, electric heating etc to maximise the benefits from DFS themselves. In terms of delivering DFS and other grid services at scale however, the first point to note is that for the majority of grid services, consumers cannot participate without being part of a collective aggregated group. This is where aggregators come in.

We also need to make things simple for the mass market. An interesting conversation that’s starting to happen at the moment in relation to the energy transition is the extent to which we need to educate consumers to make the right informed decisions in terms of energy use, versus just making it simple for them and putting the effort and investment into the infrastructure and tools that will do that for them. From a customer point of view, we all know mass market consumers only have a certain amount of time in their day to be thinking about their energy choices.

It’s a lot to be asking consumers to change their consumption behaviour to help balance the grid, however if instead they just offer the flexibility of their energy assets for aggregators to optimise, and there’s a simple interface for the customer to set preferences, that’s got to be the way to scale this. There’s a lot of different actors and players trying to get into the flexibility space. Energy suppliers and Energy Smart Appliance manufacturers are looking to do it directly and own the customer for this type of stuff. Separately, you then have new business models, such as aggregators and smart meter apps, who potentially could have a really big role in terms of engaging and owning customers and making it as simple as you possibly can for them. I think whoever makes it as simple as possible, as well as delivering the best kind of value, are going to be the winners here.

Have you been surprised about how keen people have been to be part of the DFS? The numbers of people who signed up does seem to have taken some people by surprise…

I think some people were surprised at the level of engagement in DFS, but with the high value on offer for customers to reduce demand occasionally, I wouldn’t say we were surprised and honestly think we’ve only scratched the surface in terms of consumer engagement. Octopus obviously did very well having had a head start on others after running their trial with National Grid ESO at the start of 2022. Other suppliers took a while to gear up.

What’s probably been most surprising to a lot of people, mainly the energy suppliers, is them realising that they’re not the only show in town here, and there are other routes for consumers to engage in flexibility apart from through their energy supplier. When we first started participating in the calls with National Grid on DFS and said we were going to bring consumers in through clients that are not energy suppliers, it took a long time for the suppliers to get that. We started talking about MPAN duplication issues very early on and suppliers were going “why would there be any duplication issues – customers can only be with one supplier?” – they just didn’t get it.

We’re really pleased with the number of customers that have come in through our energy supplier partners and other routes. Who knows long term who are going to be the winners here in terms of “owning” the customer for flexibility. I just think it’s good to see the innovation that is spurring on competition in the flexibility space.

I think there’s a lot of improvements that providers can implement to make it even easier and more attractive for their customers to participate in DFS and I think we’ll see significant growth in numbers for DFS next winter. We’re going to be really focusing on more automated asset participation to make it as simple as possible for consumers to participate in DFS, and soon other grid balancing programmes such as the Balancing Mechanism, frequency services and local flex services.

Have there been many barriers for either you or your non energy supplier customers signing up or participating in the trial?

There’s been a few challenges and barriers. The fundamentals of bringing customers into the service is the sharing of smart meter data. A lot of people didn’t get that this data is the customer’s data, they can share it with whoever they want provided you’ve got a proper consenting process to share that data. It’s not the energy suppliers data and that took a while to sink in.

Within SMS, our n3rgy DCC adapter service, is the biggest in the UK in terms of volumes of data coming through it outside of energy supplier’s own adapters. DCC adapters are the tool (or the plumbing I guess), used to pull smart meter data from the DCC through to wherever it should go based on the consent provided by the customer. We already had that in place at the start of the DFS service and that made it very easy to offer as part of our service to any clients that we worked with.

A lot of suppliers struggled to have the tools and processes in place to collect the data in the frequency needed to do settlements, because most suppliers are still just using smart meter data for monthly billing. But from our side we were well set up to be able to collect that data and do the baselining, forecasting, settlement runs that were needed to make the whole process work. Overall though, it’s been relatively straightforward to support all of our clients coming into the service. It was interesting to see the different types of incentive offered by different DFS Providers.

Most providers did simple cash rewards. I think there was probably a lot of frustration from some customers who felt like they had gone to the effort to turn demand down, and then see, you know, 10 pence reward coming through for an event because they had relatively low demand reduction. Shell Energy, one of our clients, went for a different approach by having a prize draw setup instead. A certain amount of customer demand reduction calculated in the event would earn them a ticket in the prize draw. A monthly prize draw then paid out larger cash rewards to lucky customers, and at the end of the DFS service, the grand prize was an Electric Mini. We actually saw higher participation from Shell customers in events than we saw from our other ‘non-automated’ clients. It may be that the prize draw model kept the engagement higher, because customers are less likely to be disappointed about a small amount of reward per event, and more interested in staying in with a chance to win a bigger prize. It will be interesting to see if this type of gamification can be a more effective way of engaging customers rather than just direct, small cash rewards.