Carbon is a terrible design metric

Sometimes the lowest footprint product is also the worst design decision you can make

Take a look at my website!

Hey Sustainability Managers,

This week, we are taking a small pause from DPP.

It hasn’t stopped being important, we will resume the conversation in the next newsletter…But after months of talking about systems, APIs, identifiers and interoperability, I realized something during a workshop last week:

A lot of teams are building incredibly sophisticated sustainability infrastructures… to optimize the wrong things.

That’s because the wrong metrics are selected and obsessing over the wrong metrics can cost you month of efforts (and budget you mught not receive next year).

For those who joined recently, I’m Gianluca. I work as a Product Sustainability and LCA expert across electronics, packaging, automotive, medtech, furniture and textiles. Which sounds very glamorous until you realize it mostly means spending large amounts of time arguing about data quality, adhesives and whether “recyclable” actually means anything in practice.

When I started doing this work, I thought the hardest part would be calculating impacts accurately.

It’s not.

The hardest part is helping teams understand which impacts actually matter for the decision they are making.

That sounds subtle, but it changes everything.

Because in product sustainability, the wrong metric doesn’t create bad reporting, it creates bad design and DPP will simply make it more evident to the common person.

Carbon became the default design metric for a very understandable reason: it’s measurable, relatively standardized, easy to communicate and finally something executives understand without needing a PhD in impact assessment.

That’s already in my opinion a miracle.

But over the last few years, carbon slowly stopped being a metric and became the metric.

And the moment that happens, teams start optimizing products around what is easiest to measure rather than what is most environmentally relevant.

You see it everywhere.

A furniture company reduces material thickness to cut embodied carbon, but suddenly the product lifespan drops from 15 years to 6 because structural integrity becomes questionable after repeated disassembly.

On paper (and marketing campaign) it is fantastic carbon reduction.

In reality you just accelerated replacement cycles, transport frequency, manufacturing demand and waste generation simultaneously. Congratulations, you optimized the spreadsheet and made the system worse.

This is the problem with isolated metrics: they flatten reality into a single number and quietly erase the trade-offs.

And ecodesign is almost entirely trade-offs.

Another common example I stumble upon is consumer electronics.

In electronics, carbon footprint is often dominated by manufacturing and semiconductor production, so teams naturally focus on reducing mass, simplifying assemblies or increasing recycled content in housings.

Which is fine until the product becomes impossible to repair.

I’ve seen devices where replacing a €3 battery requires dissolving adhesives, heating the enclosure, removing glued display layers and essentially performing surgery with the emotional pressure of a bomb disposal technician (not worth it at all at the end of the day)

The product may have a lower production footprint, but if the battery fails after one years and the device gets discarded, the environmental cost of manufacturing a replacement completely dwarfs the original optimization.

This is where durability metrics become much more interesting than carbon alone.

Not theoretical durability. Real durability.

Can components survive repeated opening cycles?
Can wear parts be accessed non-destructively?
Can firmware still function after five years?
Can the product tolerate imperfect repairs?

These questions are rarely visible in traditional LCAs because they sit at the intersection of engineering behavior and user behavior (in other words, variables that can’t be quantified precisely). But environmentally, they are often far more decisive than shaving 8% off production emissions.

Packaging has a similar problem, just inverted.

Most packaging discussions today revolve around recyclability and recycled content, which sounds reasonable until you realize both metrics are incredibly easy to manipulate.

A package can be technically recyclable and still never realistically enter a recycling stream because:

• the sorting infrastructure cannot identify it

• contamination rates are too high

• separation is economically pointless

• or the material quality degrades immediately after recovery

So when teams say:

“We moved from multilayer to monomaterial packaging”

my first question is usually:

“Did you preserve the product?”

Because packaging exists to protect something that is often environmentally far more intensive than the packaging itself.

If your beautiful recyclable pouch increases food spoilage by 4%, you may have just created a dramatically worse environmental system while improving your recyclability KPI.

And this is where packaging metrics become more nuanced.

Barrier performance. Product-to-packaging ratio. Damage rates during transport. Shelf-life extension efficiency.

Those are design metrics too. They just don’t fit nicely on marketing claims.

Textiles are perhaps the most misunderstood category of all because the industry became deeply attached to fiber narratives.

Organic cotton. Recycled polyester. Bio-based fibers.

Again: none of these are inherently wrong. But material selection without usage context is almost meaningless.

A recycled fiber garment worn seven times is environmentally catastrophic compared to a virgin fiber garment worn for ten years.

Yet longevity is still treated as an emotional concept rather than a measurable environmental parameter.

And durability in textiles is not just physical resistance.

It includes:

• repairability

• pilling resistance

• color stability

The strange thing is that fashion teams intuitively understand this already. They know which garments survive trends, repeated washing or long-term attachment.

But sustainability metrics rarely integrate these realities because they are harder to quantify than kilograms of recycled content.

So the industry keeps optimizing inputs while largely ignoring utilization.

Which is a bit like evaluating a car exclusively based on the steel composition while ignoring whether the engine works.

Furniture creates another fascinating contradiction.

A heavy hardwood chair can look terrible from a carbon perspective compared to a lightweight composite alternative.

Until you observe what happens over 25 years.

The hardwood chair gets repaired three times, survives two apartments, changes owners once, develops scratches that people strangely become emotionally attached to, and eventually gets refurbished.

The composite chair cracks near a joint after six years and quietly disappears.

Suddenly, permanence becomes an environmental metric.

Not because permanence is romantic, but because replacement has an environmental cost that most design metrics still underestimate.

This is where concepts like service life, repair cycles, modularity stability and disassembly fatigue become far more useful than isolated production footprints.

Especially under ESPR, where durability will increasingly stop being a “nice sustainability topic” and become a measurable product requirement.

What makes ecodesign difficult is that products are systems interacting with other systems:

• user behavior

• infrastructure

• maintenance

• logistics

• software

• regulation

• culture

And metrics simplify systems by design. That’s their job.

The danger starts when simplification becomes substitution.

A metric is not reality. It’s a lens.

And lenses are useful only if you understand what they distort.

The best sustainability teams I work with don’t ask:

“What’s the best metric?”

They ask:

“What behavior are we trying to incentivize through this metric?”

That changes the conversation completely. Because once you frame metrics as behavioral drivers instead of reporting outputs, you start selecting them differently.

Sometimes carbon matters most.

Sometimes durability does.

Sometimes the decisive factor is repairability, material criticality, transport density, failure probability or even software support duration.

Sometimes the most sustainable design decision is the one that looks slightly worse in the dashboard but performs dramatically better in reality, which, admittedly, can be a difficult sentence to say in a meeting full of KPI enthusiasts. But that’s also where product sustainability becomes a challenge.

Catch you next week,
Gianluca

Take a look at my website!