The Wild West of Energy Standards: Lessons from the Pre-Consolidation Stage

I recently received yet another email proposing a new energy efficiency standard from a network operator. Another one.

Today, the telecom sector is flooded with metrics claiming to measure energy efficiency. Vendors and operators frequently showcase statements like “XX% more energy efficient” in marketing materials and on oversized banners at MWC booths. But what do these claims actually mean? The truth is, the definition of energy efficiency in telecom remains unclear.

Over the past few years, equipment vendors, network operators, regulators, and organisations such as the ITU, NGMN, climate NGOs, GSMA, and even GSMA Intelligence have introduced a variety of metrics and definitions to evaluate energy efficiency and energy intensity. But why is there so much variation? And what could be done to harmonise or at least simplify these standards? Why can’t we use a single measurement, like the A-to-G rating system used for consumer appliances? 

The answer lies in three key challenges.

1. Telecom networks are complex and vast.

Unlike a simple fridge with a single energy input and an easily measurable output, telecom networks are far more complex. Networks are composed of scattered assets, often encompassing hundreds of thousands of individual components. The scope of the measurement is frequently unclear: are we measuring a single radio unit, a multiband radio, a base station, a network layer, or the entire network operator including facilities? The lack of metering further complicates things.

2. The output of the industry is dynamic.

In many applications, energy efficiency is typically calculated as useful output ÷ input. But in telecom, the output varies depending on scale. For small-scale systems, percentage efficiency might work. For macro-level systems (like buildings or national networks) energy intensity or performance indicators (e.g. per unit of GDP, product, or area) are more appropriate. Trying to fit telecom into a single output metric risks bias and oversimplification.

3. Operators compete in vastly different environments.

Imagine comparing a fridge operating in 45°C heat in the Middle East with one running at -10°C in Scandinavia. Network operators face similar disparities. Climate, geography, user density, and infrastructure maturity all affect energy performance. We must ensure operators are compared fairly.

How, then, can we consolidate energy standards into a single number like we do for cars or countries? We can’t. But here’s what we can do:

Focus on time-series comparisons like performance improvement rather than cross-sectional comparisons between networks.
Ask critical questions. “85% more energy efficient” compared to what? Per km² covered? Per GB of data? Per active connection? Was seasonality or the 35% annual data traffic growth per connection considered?
Separate trials vs. field performance. Lots of vendors report improved energy performance in a trial setting. That is welcome, but what about the results when the equipment is deployed at national scale for a mobile operator? Transparency is key here.
Use separate metrics for different layers: equipment, sites, RAN, data centres, mobile networks, and fixed networks.

 

In industries with complex systems, it has historically taken some time to standardise energy efficiency metrics and achieve scale. The same will occur in telecommunications: it is simply a matter of further harmonisation and continued effort, as the process has only begun a few years ago.  It’s always exciting when someone sets out to develop a brand-new metric. Something that’s never been measured before, using a solid and thoughtful methodology. That kind of innovation truly adds value. However, what we often need just as much is greater consistency and careful consideration, rather than simply introducing more and more benchmarks.