As the next generation of TV viewing develops in the UK and Europe, discussions are taking place about the sustainability of different distribution methods and the energy implications of the technologies used. There has been an improved understanding of the impact of media broadcast and streaming, and BBC Research & Development has contributed significantly to this.
Our collaboratively developed energy assessment methodology for broadcast and streaming is widely used across the sector, and our media lifecycle energy models of video, audio, and text are a result of in-house expertise developed over decades. Because of this research, we are well placed to examine questions about the different TV delivery methods of the future, and we’re now publishing our research that gives insight into our total UK energy footprint for video in 2024 as well as comparative broadcast and streaming energy footprints.
Viewing the entire picture – from preparation to playback
Our assessment looks at the use of energy from content preparation to distribution to consumption – but excludes production. Because of the complexity of this process, as well as the many different choices and options available to viewers, we can only understand environmental pros and cons if we assess the entire system. This ensures we understand the changes or the displacement different technological or distribution choices have on the environmental impact.
Viewing devices and consumption drive the majority of energy use
Our modelling found that around 93% of the total energy use of BBC TV and video in the UK for 2024 was mostly from viewing on devices. The distribution phase accounts for about 6% of the energy use, and the preparation of content is around 1%.
TVs continue to consume the biggest proportion of total energy, followed by set-top boxes, and personal computers (desktops) due to their higher power requirements. The digital terrestrial transmitter (DTT) network and the internet consume the most energy in distribution, and the energy from cloud services dominates the preparation phase.
It’s critical that decisions about the TV platforms of the future ensure that reductions in energy use in one part of the chain aren’t just displaced to another part of it.
Delivery makes a smaller dent in the energy use
With around 6% of the energy use resulting from the delivery of TV and video, the impact of distribution is a significantly smaller part of the overall picture. Given that different distribution methods require different hardware and devices for delivery, the overall energy consumption primarily shifts from one category of hardware to another. So, changing the delivery methods will slightly change the amount of energy used, but the same amount of consumption of video continues – just on different devices. This will result in some environmental benefits in the consumption phase – mostly driven by some viewers using smaller screen personal devices such as mobile phones or tablets instead of a TV, and a displacement of distribution energy from DTT to internet.
Overall, changing the way the video is distributed doesn’t change the total energy used by a lot, even when accounting for changes and displacements. However, the choice of using just one distribution method will have sustainability benefits over the use of several systems. It would save the energy needed to operate other distribution platforms and would reduce the total energy required to distribute video in the UK.
Streaming is catching up to broadcast in energy efficiency
As people increasingly stream content over the internet, how does the energy use compare with traditional broadcast TV? Again, it’s important to compare the entire process.
We used our modelling tools to explore purely hypothetical options to more plausible situations, and a business-as-usual scenario. Our forecast covers total and hourly energy use for the next ten years across the entire system – preparation, distribution, and consumption. The hourly energy footprints are for the total energy used, across preparation, distribution and consumption, for a representative hour of video viewed within the UK. Given the attribution involved in methodology, hourly footprints are dependent on the total audience size and viewing hours. For a business-as-usual scenario, we use Barb audience data over the last decade to estimate the shift from broadcast to streaming. With an approximately 4% year-on-year shift towards streaming based on past trends, a more aggressive audience shift in the future will lead to different results.
The average hourly energy use of online video streaming via BBC iPlayer in 2024 was around 65.7 Wh, slightly higher than the approximate 57 Wh for broadcast. The gap between the hourly energy use of traditional broadcast and the internet-based video streaming is reducing and will continue to reduce over time. That is driven by several factors, notably:
- the natural shift of audiences from broadcast to streaming,
- internet efficiency gains, delivering more data using a nearly fixed amount of energy,
- smaller screen sizes of personal devices being used more to watch streamed video.
The slight increase in the hourly energy use of broadcast (rising approximately 1.8Wh/hour over the next ten years) is driven by the shift in audiences. For broadcast, the same amount of energy being used for preparation and distribution is spread across fewer viewing hours coming from broadcast platforms as fewer viewers watch this way in the future. The shift towards less power-hungry personal devices for streaming also influences the hourly footprint.
It’s important to note the fundamental differences in the energy use of broadcast and internet infrastructure, especially as viewing changes. For example, the energy used in the digital terrestrial television (DTT) infrastructure is very similar whether one person or 25 million people are watching. So, with a decline in audiences watching via traditional broadcast platforms, the share of distribution energy used per viewing hour will increase as fewer people watch on broadcast platforms.
If things stay as they are, we expect that internet-based video streaming could become slightly more energy efficient (approximately 57.6 Wh/hour) than traditional broadcast methods (approximately 58.8 Wh/hour). Both systems could be more efficient, but that will need the use of better, more energy efficient technologies and a proactive approach for technology choices.
Better data and industry collaboration are essential
The accuracy of any model will always depend on the quality of data it uses. Even though we now have a better understanding of the media system than ever before, getting robust and reliable energy data is still difficult – especially for shared infrastructure and millions of devices used to watch video. If researchers and industry can work together, it’ll be possible to develop better energy data about how video moves from production to the viewer, reducing uncertainty and making sustainability models more accurate.