Savings Estimator
Wondering if a home battery is worth it? Our savings estimator uses backtested performance against real historical Agile rates to show you what a battery could actually earn.
Choose your battery system, consumption level, and tariff below to see a personalised estimate of annual savings and payback period.
Octopus Agile 7-Day Rate Forecast
Our ML models predict half-hourly electricity rates up to 7 days ahead. Updated daily.
What We Do
For Manufacturers
Integrate our AI trading engine into your cloud platform and make your installed base work harder.
For Developers
Open APIs for rate forecasts, dispatch signals, and consumption modelling. Build your own energy apps on our data.
For Homeowners
Free 7-day rate forecasts for Octopus Agile. Use our savings estimator to see what a battery could earn you.
How It Works
FAQs
We currently support Octopus Agile, which has half-hourly variable pricing and is the most common dynamic tariff in the UK. Support for other dynamic tariffs such as Octopus Flux, Octopus Go, and Intelligent Octopus is planned.
We use machine learning models trained on years of historical half-hourly Agile rate data, weather forecasts, and grid demand forecasts. The models predict electricity rates up to 7 days ahead. These predictions feed into our trading engine which decides the optimal charge and discharge schedule for each half-hour slot.
No. Our system works entirely through software, communicating with your battery via the manufacturer cloud API or a local API. No firmware changes, no extra hardware, no Raspberry Pi required.
We are building open APIs for rate forecasts, dispatch signals, and consumption modelling. Documentation will be available in our docs section as the API is rolled out. If you are a manufacturer or developer interested in early access, please get in touch.
Our models are trained on historical Octopus Agile half-hourly rates (from 2018 to present), weather forecast data (solar irradiance, temperature, wind speed), and grid demand forecasts. We use synthesised household energy usage patterns based on Ofgem consumption profiles for simulation and backtesting.
Community tools like Predbat are excellent but require a Home Assistant instance, a Raspberry Pi or similar device, and manual configuration and tuning. Our service is fully managed: customers activate it and it works, no local infrastructure needed. For manufacturers, we provide a turnkey integration that can be offered to their entire installed base as a subscription service.
No. A home battery works well on its own with a dynamic tariff, charging when electricity is cheap and discharging when it is expensive. Solar panels are a great addition but not required. Our system supports battery-only installations.
Yes. To benefit from automated battery optimisation, you need a tariff where the rate changes over the course of the day. Octopus Agile is the most common choice. You will also need a SMETS2 smart meter so your supplier can bill you at half-hourly rates.
Yes. Every charge and discharge cycle wears the battery slightly, and our AI factors this in automatically. Each battery model has a calibrated degradation cost per kWh cycled, derived from its chemistry, cycle life rating, and warranty terms. This cost is built directly into the reward signal the AI optimises against, so it will only cycle the battery when the expected savings outweigh the wear cost. The result is a strategy that maximises your net return over the full lifetime of the battery, not just short-term savings.