Insights

Opt-in is the failure mode: how the next decade of carbon markets has to work

2026-05-01 · Michael English · Clonmel, Co. Tipperary

Most of the carbon you emitted this year was bought from someone who never asked you a single question about it. Your electricity, your groceries, your last short-haul flight, the parcel that arrived on Tuesday — all priced, taxed, delivered, and you had no say in the climate cost. Now compare that to how voluntary carbon markets work. A button on a checkout page. A toggle on a flight booking. A small, optional, slightly embarrassing nudge that asks you, the customer, to do the right thing on top of everything else you're already paying for. We have built the entire voluntary carbon market on the assumption that people will opt in to fixing a problem they didn't opt in to causing. It was never going to work, and the numbers bear that out.

The opt-in problem is a design problem, not a willingness problem

The standard story about voluntary carbon markets is that buyers don't care enough. That's lazy. The truth is closer to the opposite — survey after survey shows people care a great deal, and a meaningful slice say they would pay more for sustainable options. Then they get to the checkout, see a tickbox that adds a few euros to a transaction they're already mentally done with, and they don't tick it. Not because they've changed their values in the last forty seconds. Because the tickbox is a tax on attention, and attention is the scarcest thing in commerce.

Every opt-in carbon offset is competing with the customer's desire to finish the transaction and close the tab. That is a fight you cannot win at scale. You can win it with a small, motivated cohort. You cannot win it with the median consumer, and the median consumer is where the emissions live.

VAT is the model, not Patagonia

Here's a thought experiment. Imagine if VAT were opt-in. Imagine if at every checkout in Ireland, you saw a little box: "Would you like to contribute 23% to public services today?" The Exchequer would collapse by Friday. We don't run consumption taxes that way because we know, as a society, that a cost which is everyone's responsibility cannot be left to individual moral effort at the moment of purchase. We bake it into the price. The buyer doesn't think about it. The seller remits it. The system works because the friction is removed.

Carbon is the same shape of problem. It is a negative externality of consumption. The honest fix is to embed the cost of mitigation into the price of the thing being consumed, so that buying a flight, a hotel night, a pair of trainers, a steak, simply costs what it costs once the climate impact is paid for. No tickbox. No moral test at checkout. No performance.

This is not a radical idea in policy circles. The EU's Carbon Border Adjustment Mechanism is a version of it. The UK ETS is a version of it. Where the voluntary market has failed is in pretending that consumer-facing offsets are a substitute for embedded pricing, when they are at best a bridge to it.

What "baked in" actually means in practice

Embedded climate cost has three properties that opt-in offsets do not.

  • It is not optional. The price is the price. The customer cannot select a cheaper, dirtier version of the same product by declining the offset.
  • It is paid by the seller out of margin, not bolted onto the buyer. This aligns incentives. The seller now has a direct reason to source lower-emission supply, because that supply costs them less to mitigate.
  • It is verifiable at the unit level. Every transaction carries a provable mitigation, not a marketing claim about a portfolio average.

That third property is where the last decade of climate finance has had its credibility problems. Bundled, opaque, off-chain registries with double-counting, phantom credits, and projects that quietly underperformed for years before journalists pulled the thread. If you are going to bake offsets into price, the offset itself has to be auditable in a way the old voluntary market simply was not.

Why we built IMPT the way we did

When we built IMPT.io, the question we kept coming back to was: what does a hotel booking look like if the climate cost is already paid by the time you click confirm? Not "would you like to add an offset", not a green badge, not a points scheme. Just — book the room, the tonne is offset, move on with your day.

So that is what we shipped. 1.7 million hotels across 195 countries. Every booking offsets one tonne of CO₂, recorded on-chain so anyone can verify it. The customer doesn't pay extra for the offset. We pay for it out of the commission we earn on the booking. That single design choice changes the conversation. The customer is not being asked to be virtuous. They are being asked to book a hotel. The climate work happens inside the price, the way it should.

We extend the same logic to the shopping side, where over twenty thousand partner brands sit inside the IMPT ecosystem, and to the IMPT Token and the IMPT Card, which let people accumulate and direct climate impact through normal economic activity rather than as a separate moral exercise. The forthcoming AI-native booking agent pushes this further again — the agent does the work of finding the room, and the offset is part of what gets booked, not a question the agent has to ask the human.

The next decade: three shifts that have to happen

If voluntary carbon markets are going to mean anything by the end of this decade, three things have to change. None of them are technical. All of them are about where the friction sits.

1. Default-on, not default-off

Every consumer-facing transaction with measurable carbon should default to mitigation included. The tickbox, if there is one at all, should be to opt out, and opting out should feel as awkward as it actually is. Behavioural economics has known this for twenty years. We applied it to pension auto-enrolment. We have not applied it to the atmosphere.

2. Unit-level proof

Sustainability claims at the company level — "we are net zero by some date" — have done more harm than good, because they have given cover to projects that did not deliver. The unit of accountability has to drop to the transaction. Every booking, every parcel, every kilowatt-hour should carry its own verifiable mitigation, ideally on a public ledger where the credit can be tracked from issuance to retirement. This is not a maximalist crypto position. It is just basic accounting hygiene applied to climate finance.

3. Seller-paid, margin-funded

The cost of mitigation has to come out of the supplier's margin, not be bolted onto the buyer's bill. This is the single change that turns offsets from a marketing surface into a procurement signal. When the airline, the hotel, the retailer is paying for the offset themselves, they start asking hard questions about where their emissions come from in the first place. That is the flywheel. Opt-in offsets never created it. Embedded offsets do.

What this is not

I am not arguing that offsets replace decarbonisation. They don't. The hierarchy still holds — avoid, reduce, then offset what remains. What I am arguing is that the voluntary market, as currently structured around opt-in consumer choice, is a poor mechanism for funding the residual mitigation we will need for decades while the harder work of grid decarbonisation, aviation fuels, cement, and steel grinds through. If we want climate finance to scale, we have to stop relying on the conscience of a tired person at a checkout.

The cleanest version of this is regulatory — a proper price on carbon, applied across borders, paid at source. The next-cleanest is voluntary embedding, where platforms and brands bake the cost in themselves rather than wait for the policy to catch up. Both can run at the same time. Neither requires a single additional tickbox.

What to do this week

If you run a business that sells anything with a carbon footprint, do one thing this week: get a number on your average per-transaction emissions, and then get a quote on what it would cost you, out of your own margin, to mitigate that automatically on every sale. You will probably find the number is smaller than you feared and the operational lift is smaller still. That single exercise is what moved us from talking about climate to shipping a platform where 1.7 million hotels offset a tonne per booking without asking the guest to be a hero. If you want to see how the embedded model looks in production, the platform is live at IMPT.io. And if you would rather just argue with me about it, I am in Clonmel and I read my own inbox.

Reading these regularly?

Subscribe to Letters from Clonmel — quarterly long-form founder letters from Mike. First letter Q2 2026.

Subscribe →