Web3 has been promising to deliver for years real possessions on the chain. In carbon markets, that promise is starting to materialize, but not for the reasons most people expected.
The first story was simple: blockchain would make carbon credits transparent, tradable and reliable. Tokenize the credit, put it in a ledger and the problem is solved.
That narrative has been repeatedly tested since 2020. It didn’t last.
What is changing carbon markets now is not just tokenization. They are improvements in monitoring, reporting and verification (MRV) – especially digital MRV (dMRV) – that aim to make carbon results more observable, more frequent and more auditable. Distributed ledgers still have a role to play, but increasingly as infrastructure on top of better data, not as a replacement for it.
Carbon 1.0: why the first wave struggled
Between 2020 and 2024, the voluntary carbon market faced a credibility crisis. Research and academic studies have raised concerns about over-crediting, weak baselines, non-additionality and persistent risks in parts of the market. These issues did not apply uniformly, but were widespread enough to impact buyer confidence.
A key limitation was upstream: how the carbon results were measured and verified.
Traditional MRV processes often relied on:
As a result, many credits represented retrospective estimates rather than near-real-time measurements.
Tokenization did not solve this. Recording a credit on a blockchain can improve traceability and auditability, but it does not improve the underlying data. If the input is uncertain, the output remains uncertain – only more visible.
Price signals begin to differentiate quality
Between 2025 and 2026, markets have begun to differentiate more clearly between credit types, in particular between avoidance-based credit and carbon dioxide removal (CDR).
-
Developed and highly sustainable CDR credits (e.g. DAC, mineralization) have been reported in the range of $170-$500+ per tonne depending on method and contract structure.
-
In some data sets, Biochar credits are reported in the mid-$100s per tonne, although prices vary widely between suppliers and deal structures.
-
Higher rated credits (for example those that score well under independent rating frameworks) account for a growing share of total pension value, although lower quality offerings continue to represent a significant portion of the total market volume.
These are still thin and fragmented markets, but the direction is clear: data quality and sustainability are increasingly reflected in price.
What dMRV actually changes
Digital MRV refers to the use of remote sensing, automation and digital data pipelines to improve the way carbon results are measured and verified.
Rather than relying solely on periodic audits, dMRV systems can include:
-
satellite images,
-
LiDAR and geospatial analysis,
-
IoT sensors,
-
and automated data ingestion into verification systems.
This does not always mean continuous real-time measurement, but it can significantly increase data frequency, reduce reporting delays and improve auditability.
Conceptually, this shifts carbon credits from static certificates to data-backed, periodically updated records.
The architecture generally consists of three layers:
-
Measuring layer
Providers of remote sensing, sensors, and project-level instrumentation generate observation data. -
Verification layer
Statistical models, machine learning and rule-based systems are used to track changes (e.g. deforestation, shifts in biomass) and flag anomalies. These systems still require validation and are not error-free. -
Registry/infrastructure layer
Registers and, in some cases, distributed ledgers record issuances, transfers, and retirements. At this stage, infrastructure becomes more useful because it is anchored in higher quality inputs.
A real milestone: Verra’s dMRV pilot
In February 2026, Verra approved the first credits under a dMRV pilot.
The initial project:
-
is a solar installation in the Comoros,
-
uses fully digital data delivery and verification,
-
allows monthly or bimonthly issuance instead of annual cycles,
-
and includes a security structure where 80% of credits are issued initially and 20% are retained pending further validation.
This is a meaningful milestone, but it is explicitly a pilot and not a market-wide shift.
More complex project types, such as large-scale forestry or some forms of DAC, still face challenges, including:
-
costs of instrumentation,
-
unequal data availability in different regions,
-
and the difficulty of validating models in ecosystems.
The implication is not that MRV has been ‘solved’, but that a new direction is being tested in production environments.
Towards more programmable carbon workflows
One area where improved data could have a downstream impact is in the way carbon credits are used operationally.
Today, most compensation still takes place during an annual accounting cycle. Emissions are calculated afterwards and credits are purchased and withdrawn accordingly.
At the same time, some major buyers are experimenting with higher resolution energy matching:
-
Google aims to operate 24/7 on CO2-free energy by 2030.
-
Microsoft has piloted hourly rate matching methods in specific contexts and continues to expand its carbon removal tenders.
In addition, companies including Microsoft have signed long-term offtake agreements for technical moves, including projects such as STRATOS direct air capture plant, which has scheduled first operations around 2026.
These developments suggest a possible future in which:
-
emission data is generated more often,
-
the carbon supply is fixed in advance,
-
and parts of the purchasing and retirement process could be automated.
Technologies such as smart contracts could support this type of automation. However, fully automated, real-time carbon matching systems are still in development and have not yet been widely deployed.
Carbon as an emerging financial asset
Carbon credits are increasingly evaluated using tools well known in the financial markets:
-
Independent companies such as Sylvera provide quality ratings (AAA to D scale) that assess integrity and supply risk.
-
Higher rated credits are often associated with price increases and stronger buyer demand.
-
Standardization efforts, including the Integrity Council for the Voluntary Carbon Market Core Carbon Principles (CCPs), aim to define minimum quality thresholds for the market.
That said, carbon is not yet a fully mature asset class:
Tokenization and digital settlement infrastructure are being explored by both startups and incumbents, including pilots involving major financial institutions. These efforts can improve settlement speed, transparency, and interoperability, but they are still evolving.
Regulations and the “Integrity Premium”
The policy also reinforces the importance of reliable carbon data.
The European Union’s Carbon Border Adjustment Mechanism (CBAM) will enter its final phase in 2026. Importers of certain goods must account for embedded emissions and purchase corresponding certificates, making carbon accounting a matter of trade compliance rather than voluntary disclosure.
At the same time, voluntary market initiatives such as CCP labels are beginning to influence the purchasing decisions of major buyers.
A consistent pattern emerges in both compliance and voluntary systems:
more confident data → stronger demand → higher prices
But this ‘integrity premium’ is still emerging and varies considerably by market and methodology.
Risks that do not disappear
Improved infrastructure changes where risks exist; it does not eliminate them.
The main risks include:
-
old supply: large amounts of older credits with weaker methodologies are still circulating
-
measurement risk: sensors can malfunction or be manipulated; remote sensing has limits
-
model risk: AI and statistical models may not generalize well across geographic areas or ecosystems
-
standard fragmentation: multiple registries and methodologies remain in use
-
Financialization risk: as markets mature, there is a risk of optimizing on price or liquidity rather than climate impact
These limitations are critical to how the market will develop over the next decade.
Conclusion
Tokenization alone has not solved carbon markets. Improvements in measurement and verification are beginning to address some of the underlying challenges.
dMRV does not make carbon perfectly measurable, but it can make it more observable, timely and controllable. That in turn makes the downstream infrastructure (be it registries, marketplaces or programmable settlements) more meaningful.
Carbon markets have not yet been fully transformed. But they’re starting to shift:
from static claims → to data-based environmental assets
Whether that shift will happen at scale will depend less on the infrastructure stories – and more on whether the underlying data continues to improve.