What can a company honestly claim about its sustainability ethics?

This page treats sustainability ethics as practical questions: what can be claimed, what can be measured, and what to do when environmental goals conflict with each other or with something else that matters.

One thing to be clear about before the questions: a large share of published sustainability information cannot be independently checked by the people reading it. That is not an accusation. It is a structural feature of how the data is produced, and any honest treatment of the ethics has to start there.

What to take away

Most published sustainability figures are estimates rather than measurements, so two honest companies with identical operations can publish different numbers and both be defensible.
State the basis of every figuremeasured, estimated, modeled, or reported by someone else. This one habit does more for credibility than any narrative.
A certification tells you a scheme's criteria were met on the day it was assessed. It does not tell you the criteria are the ones that matter to your reader.
Where an environmental goal and a social one conflict, name who bears each cost before choosing. The conflict is usually real rather than a communications problem.

Why most of it cannot be checked

Because most of the numbers are not measurements.

From meter to estimate

Metered: direct energy at owned site
Estimated: activity data x conversion factor
Estimated: spend x intensity factor
Estimated: supplier's own estimate passed upward
Further from operations = more modeling

Direct energy use at a site you own can be metered. Almost everything beyond that is estimated: activity data multiplied by a conversion factor, or a spend figure multiplied by an intensity factor, or a supplier's own estimate passed upward. The further from your own operations a figure describes, the more of it is modeling.

That produces several honest consequences:

Two companies with identical operations can publish different figures by choosing different factors, boundaries, or baselines, and both can be defensible. How such claims may be phrased to consumers is separately regulated in some markets, and the FTC's green guides set out one regulator's position on environmental marketing claims.
Year-on-year changes can be dominated by a methodology revision rather than by anything happening in the world.
A number carried to three significant figures is conveying a precision that the inputs do not support.
Assurance, where it is obtained, often covers whether a process was followed rather than whether the number is close to true.

The ethical obligation this creates is not to stop reporting. It is to stop presenting estimates in the grammar of measurements. Say which is which. That single habit separates the organizations worth believing from the rest, and it costs nothing.

Claims you can actually support

The test is whether the sentence would survive someone competent asking three questions: what was measured, over what boundary, against what baseline.

Three questions before publishing

What was measured?
Over what boundary?
Against what baseline?
Could someone reproduce it from the data?
If not, narrow the claim

Claims that generally hold up:

A specific change, at a named boundary, against a stated baseline, by a stated method. "Energy use at these four sites fell by this much against this year, metered."
A description of an action taken, without an outcome attached. "We changed this material in this product." Actions are verifiable; benefits usually are not.
A commitment with a date, a scope, and an accountable person.

Claims that do not hold up:

Any absolute adjective (clean, green, sustainable, eco), applied to a product with a material footprint. These describe a direction at best and are read as a state.
Net figures where the netting is doing the work, unless what was netted is stated in the same sentence.
Comparisons against an unnamed alternative. Better than what?
A whole-company claim supported by evidence from part of the company.
Anything about your supply chain that rests entirely on a supplier's own declaration, presented as though you had established it.

The reliable internal test: could someone reproduce this claim from the data if you handed it to them, and would they get roughly what you published? If not, narrow the claim until they would.

Claims about environmental performance are regulated as consumer or investor communications in a growing number of markets, and the rules differ by jurisdiction. Whether a particular phrasing is permissible where you sell is a question for counsel, not for the sustainability team alone. In the United States, Section 5 of the Federal Trade Commission Act prohibits unfair or deceptive acts or practices; the FTC's Green Guides explain how the agency applies that standard to environmental marketing claims.

Certifications and labels

Treat any scheme as a claim that needs the same scrutiny as your own. Examples include B Corp certification (administered by B Lab), Forest Stewardship Council (FSC) certification, Fairtrade and Rainforest Alliance labels, and ISO 14001 certification for environmental management systems. The Science Based Targets initiative (SBTi) validates corporate climate targets, but it is not a product ecolabel. These schemes cover different scopes; a mark alone does not establish that an entire company or product is sustainable.

Questions worth asking before relying on one, or before putting its mark on a product:

What exactly is certified?Frequently it is one input, one site, or one attribute, not the product a customer is holding.
Who verifies, and who pays them?A scheme where the certified party selects and pays the verifier has the same structural pull as any audit.
Is it a standard or a membership?Some marks indicate an assessed threshold. Others indicate participation in a program. Customers cannot tell these apart, which means you are relying on their confusion if you know the difference and do not say.
How is chain of custody handled?Mass balance and book-and-claim systems are legitimate accounting approaches, but they mean the certified material may not be in the item bearing the mark. If your customers would be surprised by that, saying so is part of the claim.
What happens when a certified party fails?A scheme with no observable enforcement is a subscription.

None of this makes certification useless. It makes it evidence of a specific, limited thing, and the ethical failure is describing it as evidence of something broader.

For U.S. consumer-facing claims, the Federal Trade Commission and state attorneys general are relevant enforcement authorities. Their oversight of advertising claims is separate from a certifier's own complaint, suspension, and appeal procedures.

Offsets, honestly

The defensible position is narrow, and it depends on what you are claiming with them rather than on whether you buy them.

Using purchased credits to fund work you think worthwhile, and saying so, is straightforward.

Claiming they cancel your operations or product's climate effect is trouble.

That claim rests on counterfactual assumptions: the avoided emissions would otherwise have occurred, they will not move elsewhere, and the storage lasts as long as the emission does. Buyers cannot verify this, and whole credit categories are contested.

Examples of offset registries and crediting standards include the Verra Registry, the Gold Standard Impact Registry, Verra's Verified Carbon Standard (VCS), the Gold Standard, and the UN Clean Development Mechanism (CDM). These names identify programs and records; they do not by themselves prove that a particular credit is additional, permanent, or matched to the emissions in a claim.

Practical consequences:

Keep purchases and reductions in separate lines. Combining them hides the number that matters.
Do not let a credit purchase substitute for a reduction you could actually make. If the offset is cheaper than the fix, that is a fact about credit pricing, not about the right thing to do.
Be specific about permanence and timing when you say anything at all. An emission today and a removal over decades are not the same quantity.
Rules on offset-based claims differ by market. The EU's Empowering Consumers for the Green Transition Directive restricts certain product claims of neutral, reduced, or positive environmental impact when they are based on greenhouse-gas offsetting. In the United States, consult the FTC Green Guides and take legal advice on the wording before publishing.

When two environmental goals conflict

Frequently, and this is the part that generic guidance skips.

A lower-carbon material may use more water, come from a place where extraction damages a habitat, or be harder to recycle, while a more efficient facility may consume land. Making a product last longer may raise its manufacturing footprint.

Local sourcing cuts transport, often a small share of total impact, but can raise production impact.

There is no formula that resolves these, and the ones that appear to, single-score environmental indices, resolve them by embedding weightings that someone chose and that are usually invisible to the person reading the score. The Environmental Performance Index (EPI) is one named composite environmental ranking, but it ranks countries rather than companies or products; company decisions still require disclosure of their own scope and weighting.

What to do instead:

Show the trade-off rather than collapsing it.Present the dimensions separately. A decision-maker who sees two numbers moving in opposite directions makes a better decision than one who sees a single index.
State the weighting explicitly when you must choose."We prioritized emissions over water use because of where this facility sits" is a defensible sentence. A score that silently does the same thing is not.
Watch the boundary.Most apparent wins are boundary effects: the impact moved outside what you count. Ask where it went.
Do not compare against the ideal.The relevant comparison is the realistic alternative, including doing nothing.

Environmental against social

In the United States, the California Transparency in Supply Chains Act requires covered retailers and manufacturers to disclose their efforts to eradicate slavery and human trafficking from their supply chains. It is a disclosure law, not proof that a company's suppliers are free of abuse.

This is harder than the previous question and much less discussed.

Closing a high-emitting operation removes emissions and a workforce's livelihood; efficiency programs eliminate jobs. Lower-carbon technology requires materials extracted under conditions that would fail your labor standards.

Raising standards for suppliers in a low-income region may push production to somewhere with weaker oversight. A company facing that trade-off is making a corporate responsibility decision about effects it causes, not a communications one.

Some honest positions:

The costs of an environmental improvement usually fall on people other than the ones who benefit, and usually on people with less power. That is a fact about the decision, and it belongs in the paper.
"Just transition" is a real idea that is easy to say and expensive to do. If you use the phrase, it should be attached to specific money and specific commitments to specific people, with dates.
Where you cannot resolve the conflict, say which one you chose and why. The alternative, reporting the environmental gain and not the social cost, is the standard practice and it is a form of dishonesty even when every individual number is accurate.

Selling a product with a large footprint

Selling something with an environmental cost is not in itself a wrong. Nearly everything has one.

The questions that matter are narrower:

Are you describing the product accurately, including what it costs environmentally?
Are you making it as good as you reasonably can, or only as good as it needs to be to carry the marketing?
Are you working against measures that would reduce the harm (through lobbying, industry association positions, or political contributions), while publishing a commitment to the opposite?

The largest gap in most corporate sustainability positions is between what a company publishes and what it funds. Trade association memberships and political spending are where behavior and stated positions diverge most often, and where that divergence is least reported.

If you check one thing inside your own organization, check that.

Common questions

What should we measure, and what should we not target? Measurement is worth doing. Attaching bonuses to it needs care, because any environmental metric with money attached will be managed.

Prefer measures close to what you control.Metered energy, waste to landfill from your sites, water drawn. These can be verified and improved by actual work.
Be careful with intensity metrics.Emissions per unit of revenue can improve while absolute emissions rise. Both are legitimate; publishing only the flattering one is not.
Watch for boundary management.Divesting a high-emitting business reduces your reported footprint and changes nothing in the world. If your figures improved because of a transaction, say so in the same line.
Do not target a supplier-reported number.You will get better reported numbers.
Keep one absolute figure as the headlineeven when it is the least flattering, because it is the only one that cannot be improved by restructuring the denominator.

What is our obligation if we discover a past claim was wrong? Correct it, in the same place and with the same prominence, and say what changed.

The temptation is to fix it quietly in the next report: restate the baseline, adjust the figure, and let the correction sit unremarked in a footnote. This is common enough to be near-standard, and it is the practice most corrosive to the credibility of the whole field, because it means historical figures can never be relied on.

If the wrong claim was made to customers or investors, the handling of the correction may have legal dimensions that go beyond editorial judgment. Take advice on the sequence and the wording before publishing, but take it on the basis that you are correcting it, not on whether to.

Where should a small organization start? Don't try to adopt every framework at once. Common references include GRI Standards, IFRS Sustainability Disclosure Standards issued by the International Sustainability Standards Board (ISSB), Task Force on Climate-related Financial Disclosures (TCFD) recommendations, and CDP's disclosure system. Choose what fits your reporting obligations and audience, then do four things:

Meter what you can meter. Actual consumption at your own sites, actual waste weights. Real numbers for a small boundary beat estimates for a large one.
Find your largest single impact honestly, even if it is one you cannot currently do anything about. Most organizations are surprised, and most published programs address the second or third largest because the first is inconvenient.
Make one change that costs something. It establishes whether the organization can actually do this.
Publish nothing you cannot evidence. A short, dull, accurate statement ages far better than an ambitious one, and it does not become a liability when the standards tighten.

Related reading on this site: customer ethics covers the same honesty problem in what you tell buyers, and leadership ethics covers who decides when the honest claim is the smaller one. Corporate responsibility covers the wider question of what a company owes for its effects, while supplier ethics covers the part of your footprint inside somebody else's operation.