Sustainability teams tend to meet the e-waste environmental impact question at the end of the asset lifecycle, framed as a diversion metric — tonnes kept out of landfill, reported annually, filed under waste. That framing is convenient and it undercounts the number badly. By the time a device reaches disposal, the overwhelming majority of its environmental cost has already been paid, and diversion tonnage doesn’t measure it at all.
Most of the footprint arrives before the device does
For a laptop, manufacturing typically accounts for the large majority of lifetime emissions. Operational electricity — the part organisations actually meter and report — is the minority share.
The reason is the supply chain sitting behind every component. Silicon wafer fabrication is one of the most energy-intensive industrial processes in existence. Rare earth separation runs on solvent extraction across dozens of stages. Then there’s global freight, several times over, as subassemblies move between countries before final assembly.
None of that appears in a facilities energy bill. It’s embodied, it’s already spent, and it’s the majority of the number.
Ore grade is the argument for urban mining
Here’s the comparison that reframes the whole discussion.
Primary gold mining now works ores measured in grams per tonne. Move enormous volumes of rock, use enormous volumes of water, produce tailings, recover a few grams.
Printed circuit boards from retired IT equipment carry gold concentrations orders of magnitude higher than the ore currently being mined for it. The same holds for copper, palladium, and silver. A pallet of decommissioned servers is a richer deposit than most active mine faces.
That’s the case for treating a decommissioning project as a resource stream rather than a waste stream. The concentration work has already been done — by the manufacturer, at their expense, using their energy.
Diversion tonnage measures the wrong thing
A diversion figure treats a working laptop and a shattered monitor as equivalent inputs, weighted by mass. It rewards throughput.
Two organisations can report identical diversion tonnage while producing completely different outcomes. One shredded everything for material recovery. The other refurbished 60% of it into another three to five years of service and recovered material only from what genuinely failed.
The second organisation avoided the manufacturing footprint of every device it kept in circulation. That avoidance is the largest single environmental lever in the entire lifecycle, and a diversion metric renders it invisible.
What to measure instead
Report reuse rate — the share of decommissioned units returned to productive service, whether internally, through resale, or through donation. This is the number that tracks avoided manufacturing.
Report average service life at retirement, and track it over time. A refresh cycle stretched from three years to four cuts the annualised embodied footprint of the fleet by roughly a quarter, and it does so without any change to disposal practice.
Report placement outcomes with named recipients where donation is involved. Verifiable specifics are what distinguish a sustainability disclosure from a marketing claim.
Keep the diversion number. Just stop leading with it.
See also: The Role of WhatsApp Page and Telegram Chinese in Hong Kong’s Tech Savvy Culture
Procurement is where the lever actually sits
If most of the footprint is embodied in manufacturing, the highest-impact decisions happen at purchase, not disposal.
Buy for repairability. A device with a replaceable battery and accessible storage will stay in service years longer than a glued-shut equivalent, and battery failure is the single most common reason serviceable hardware gets retired early.
Extend the standard refresh interval wherever the workload allows it. Most office roles have not been genuinely constrained by four-year-old hardware for a decade.
Buy refurbished for suitable roles. Nothing beats a device whose manufacturing footprint was paid for by someone else.
Reframing the e-waste environmental impact metric
The e-waste environmental impact of an IT estate is decided by how long devices stay in service and how completely they’re recovered afterwards — in that order of importance. Disposal practice matters, but it’s the smaller half of the equation and it’s the half most reports treat as the whole.
Global Recycling Day on March 18 is a reasonable annual prompt to revisit this. The more useful exercise is running the reuse-rate number alongside the diversion number for a year and seeing which one your programme is actually optimising for.
