Reduce your heating costs by letting Tibber control your heating smarter!
With a connected thermostat, heat pump/-boiler or radiator you can easily set the desired temperature in the Tibber-app. The heating device will then be controlled automatically by our smart algorithms which take things like weather forecasts and electricity prices into account when making decisions about how and when to heat.

Heating can be done in a bunch of ways but the goal is to heat the air in your home and to keep that heat as long as possible. This is usually done by:

Most of these devices have one thing in common: They all use target values (also called setpoint) of some sort, together with measured changes in the indoor temperature to know when to heat, how much and when not to heat.

Tibber collects information regarding indoor temperature, target value, and running mode from your device. With that we create a virtual model of your device and add information from the weather forecast, desired temperature and electricity prices. All this is then analyzed by our machine learning algorithm to find correlations and patterns within this information. This gives us insights into things like your home’s heat inertia.
​

The machine learning algorithm that continuously studies your device and the input, will take all mentioned 👆 into account when it creates a schedule of how to heat your home for the coming hours. This is done one to four times per hour depending on the type of heating device it is controlling. A slower-moving system (e.g. ground source heat pumps) will be updated less frequently than a faster-moving system (e.g. air-to-air heat pumps).

We communicate with the device by sending commands to the server (API) of the company that has made the device that we have integrated with. These commands will then be communicated to your specific heating device, either directly or through another device such as a Sensibo Sky or Air, which in turn forwards the commands to the heat pump with IR-signals, just like a remote control does.

Some devices we control by changing the target value directly while we for other devices move the heat curve up and down. There are also examples where we set schedules in the manufacturers native app or server that deploys the planned changes when the time comes. But all these devices have one thing in common: When the machine learning algorithm has run and an update needs to happen, we’ll send a new command to the servers (API) of the company that has made the device.

This means that if we set a target value that is of a higher temperature than the measured indoor temperature, your device should start working to produce and/or distribute heat.
When setting a target value that is of a lower temperature than the measured indoor temperature the device should stop heating/distributing heat while waiting for the measured indoor temperature to drop below this value, until it starts heating again.
Please note that depending on the setting and the device type, air may still be circulated. For example when using an air-to-air heat pump.

If you usually spend your days at an office and your nights asleep in bed there’s no need to keep your home at normal temperature. So, with these settings activated we lower the temperature of your home by approximately 3 degrees while you are away or sleep and heat it back up in time for you to wake up or return back home. Smart right?

If you have this mode activated, we’ll try to keep the temperature in your home at around 8-10 degrees (if your device supports such a low setting through their API). There’s no need to have it fully heated when no one is home, right?

This works in the same way as smart heating does when you have that set at the same desired temperature, or as when you control your device manually, depending on the device type. It differs between devices. For some, it works no matter if you have smart heating activated or not, for some we can’t activate it at all.
​
Read more about it here.

We control these by changing the target value. If the device is in any other mode than in “heat” it will actively cool air when we change the target temperature to one of a lower value than the measured indoor temperature. This means it won’t save you any energy.
If you instead run the device in “heat mode” it will stop producing heat when we change the target temperature to one of a lower value than the measured indoor temperature, and thereby use less energy.

Yes, as we’ve already mentioned we use electricity price as an input when making decisions about when to, how to heat and when not to heat. This lets us “move” energy consumption between hours. This means we might heat more on lower priced hours and less on more expensive hours. So, if your home is heated by anything that uses something else than electricity (that is priced according to the spot price of your area) for its power source, like gas, district heating, or fire, you should not turn smart heating on.

Equivalent or similar functions in the native app, to the ones that Tibber provides, should not be used when you want Tibber to control the heating, as it may compete or be overwritten when used simultaneously. Examples of such functions are scheduling, switching off and on, and adjusting modes or controlling temperatures. You should therefore choose an app and its functions and only use these.

If you for example have smart control on and change the temperature in the native app, there is a risk that we do not catch the change, and thus overwrite that setting when our algorithm has analyzed the information and made an updated decision.
​
You can find our FAQ about smart heating here.