CBAM covers six sectors, and the rules for calculating embedded emissions differ enough between them that “does CBAM apply to me” only gets you halfway — you also need to know which emissions actually count.

In this article:

• A single comparison table covering all six CBAM sectors side by side
• Why some sectors count electricity emissions and others don’t
• The one biggest emissions-reduction lever in each sector
• A quick sector-by-sector rundown of covered products and standout numbers
• Links to a full deep-dive post on each sector

I get some version of the same question from almost every operations or finance leader I talk to about CBAM: “does this apply to us, and if so, how bad is it?” The honest answer depends enormously on which of the six covered sectors you’re in — cement, iron and steel, aluminium, fertilisers, hydrogen, and electricity — because the underlying rules for what counts as “embedded emissions” genuinely differ from one sector to the next.

That’s not a minor technicality; it changes both the size of your CBAM bill and where decarbonisation spending pays off first. This post is the hub for our six-part sector series: one table, six short sections, and links to a full deep-dive on each sector.

CBAM’s Six Sectors at a Glance

Here’s the comparison that most readers actually want. “Direct-only” means only emissions from the production process itself count toward the CBAM bill; “direct + indirect” means emissions from the electricity used to make the good count too.

A grouped view of which sectors count electricity (indirect) emissions and which don’t — most readers assume it’s all six; it’s really two.

* Sintered ore, one precursor within the iron & steel category, is the exception — it still counts indirect emissions.

If you take one thing from that table, take this: most people assume CBAM counts the full carbon footprint of everything it touches, electricity included. It doesn’t. Four of the six sectors sit on Annex II to the CBAM Regulation, the list of goods where only direct emissions count. Only cement and fertilisers count both.

Cement: Where the Carbon Comes From the Chemistry, Not the Kiln Fire

CBAM’s cement category covers clinker, calcined clay, ordinary and blended cement, and aluminous cement. Counter-intuitively, most of a tonne of cement’s embedded emissions don’t come from burning fuel to heat the kiln — they come from the chemistry itself. Heating limestone to make clinker releases CO2 as the carbonate breaks down, and that calcination process is the single largest emissions source in the sector, ahead of combustion.

Cement is one of only two sectors where indirect (electricity) emissions count toward the CBAM number. The standard default factor for clinker is 0.525 tonnes of CO2 per tonne. Because clinker drives the bulk of embedded emissions, the biggest lever is the clinker-to-cement ratio: ordinary Portland cement typically runs 80–95% clinker, while blended cements permitted under European standards can run as low as 5–20%, substituting in gypsum, slag, fly ash, or calcined clay instead — a massive potential swing for what a buyer might otherwise treat as an interchangeable product.

Iron and Steel: The Widest Product List, and a Route Choice That Matters More Than Geography

This is the broadest CBAM sector by far. It stretches from sintered ore and pig iron through crude steel all the way down to finished pipes, structural beams, and even screws and bolts. If you import fabricated steel goods into the EU, you’re very likely in scope even if you’ve never thought of yourself as a “steel importer.”

Production route matters enormously here. The traditional blast-furnace route (BF-BOF) uses coke as a reducing agent and is structurally carbon-heavy; the electric arc furnace route (EAF), which mainly melts scrap using electricity, is structurally lower-carbon — scrap itself is treated as having zero embedded emissions. Default benchmark values reflect that gap: roughly 1.364 tonnes of CO2 per tonne for BF-BOF crude steel versus roughly 1.189 for EAF/stainless steel, even stacked with carbon-intensive alloy inputs. Nearly the entire category counts direct emissions only — sintered ore is the lone exception. The biggest lever is the choice of route: shifting from blast furnace to electric arc production, or sourcing from a supplier who already has.

Aluminium: The Sector Where “Direct-Only” Really Changes the Math

Aluminium covers unwrought aluminium (ingots, billets, slabs) and a wide range of downstream products — bars, wire, sheet, foil, tube, structural pieces. Its real-world carbon footprint is dominated by the electricity consumed in electrolysis; smelting a tonne of primary aluminium is one of the most electricity-intensive industrial processes there is. But aluminium sits on Annex II, so only direct emissions count — that huge electricity footprint is, for now, outside the CBAM calculation entirely.

What does count directly: carbon anode consumption during electrolysis (the dominant source) and PFC gases released during rare “anode effect” upsets in the electrolytic cell. Those PFC gases matter disproportionately because their global-warming potential is thousands of times that of CO2. The single biggest lever is production route, not electricity source: secondary (recycled) aluminium has no anode consumption and produces no PFCs, so it shows structurally lower embedded emissions than primary aluminium regardless of where it’s made.

Fertilisers: Ammonia Is the Hub Almost Everything Else Runs Through

CBAM’s fertiliser category covers nitrogen-containing products only — ammonia, nitric acid, urea, and mixed N-P-K fertilisers. Phosphorus-and-potassium-only blends are excluded because they carry no significant embedded emissions.

Ammonia is the structural center of this sector: it’s a direct precursor for nitric acid, urea, and mixed fertilisers, so its carbon intensity ripples through nearly the entire product family. Ammonia synthesis is fuel-intensive (mostly steam-reformed natural gas), while nitric acid has its own distinct issue — process nitrous oxide (N₂O) from catalytic oxidation, which matters far more than its mass suggests because N₂O’s global-warming potential is 265 times that of CO2. Fertilisers count both direct and indirect emissions. The biggest lever is decarbonising ammonia production — cleaner hydrogen feedstock, more efficient reforming, or carbon capture — since that one input cascades into nearly everything downstream.

Hydrogen: Simple Math, But the Production Route Changes Everything

Hydrogen is the narrowest CBAM sector — a single CN code covering pure hydrogen. It’s also structurally the simplest to calculate, since hydrogen has no CBAM precursors of its own; its embedded emissions come entirely from its own production process.

Here’s the genuinely counter-intuitive part: hydrogen is direct-emissions-only, meaning the electricity used to make it via electrolysis doesn’t enter the calculation at all. So the “grey/blue/green” labels used in the hydrogen industry don’t map cleanly onto the CBAM number — a grid-powered electrolysis producer and a renewables-powered one can show similarly low direct emissions, while steam-reforming producers face the full weight of their fossil combustion emissions, uncapped. The biggest lever is production route: switching from steam reforming to electrolysis, or adding carbon capture, has an outsized effect precisely because only direct emissions count.

Electricity: A Sector Defined by Geography, Not Trade Volume

Electricity is a different kind of CBAM category altogether — not a manufactured product with a CN code, but the commodity itself, imported as cross-border power flows onto the EU grid. Only direct emissions count, calculated using an emission factor tied to the exporting country or region’s generation mix; defaults apply unless very specific criteria are met.

Using actual, installation-specific emissions instead of a default requires clearing a high bar: a direct power purchase agreement with the specific generator, no grid congestion, emissions under 550 grams of CO2 per kilowatt-hour, capacity nominated and matched to production within a one-hour window, and monthly verification. Because electricity can only flow through physical grid interconnections, this sector is inherently regional — it applies to the EU’s directly connected neighbours, not distant trading partners the way the other five sectors do. The biggest lever is straightforward: a cleaner generation mix at the specific plant supplying the EU grid.

How Sprih Helps

Once you know which sector you’re in and whether indirect emissions apply, the real work starts: getting accurate embedded-emissions data out of your suppliers, tier after tier, for every precursor that feeds into what you import. That’s a supply-chain data problem before it’s a compliance problem, and it looks different in every sector — tracking clinker ratios in cement, production routes in steel and aluminium, or precursor chains in fertilisers all require different inputs from different people.

Sprih’s core product, SustainSense, is an AI agent layer that indexes data from more than 120,000 companies and 400,000+ sustainability reports — think of it as AI infrastructure for climate, the way Bloomberg is infrastructure for finance. We’re building CBAM-specific tooling on top of that foundation to help importers and their suppliers calculate and manage embedded-emissions data across all six sectors, so the sector-specific nuances above don’t have to be tracked by hand. Companies that treat CBAM as a supply-chain data problem — rather than a once-a-year filing exercise — tend to turn it into a genuine competitive advantage over slower-moving competitors.

Sprih is an AI-native sustainability intelligence company helping enterprises turn climate and compliance data into competitive advantage.