Methodology

How the register fits together

Everything hangs off deployments: a product, service or programme running on a ledger for real users. Each one names its participants (including the firm whose platform it is built on), the networks it settles on, the assets it issues or uses, what it is used for, where it operates and the rules it operates under. Click a box to browse that part of the register.

What counts as a deployment

One named product, programme or service run on a ledger for real users, with its own operator or issuer, participants, status and measurable activity. It is not the software: the firm whose platform it is built on is recorded as a participant with the platform role, and deployments sharing a branded platform carry its name.

1. No source, no claim

Each fact in a record (a participant, a network, a use case, a status, a figure) carries the id of the source behind it. The build fails if any claim is unsourced, if a reference points at something that doesn't exist, or if a record isn't in the expected shape. Summaries are written in our own words.

Coverage is global. If something isn't in the register, we haven't found it yet; it isn't out of scope. We use sources in any language, because a local regulator or operator is often the best primary source. Where a source isn't in English, the source list says so and our summary is a paraphrased translation.

2. The status ladder

Where a deployment is today. Status only moves when a dated, sourced event says so; history is kept, never overwritten.

Announced

Public intent from a named organisation. No transactions yet.

Proof of concept

A proof of concept: test environment, simulated value, or a one-off demonstration.

Pilot

Real value moves, but with a handful of participants or in controlled production.

Live (limited)

In production with real clients, but scope, eligibility or volume is restricted.

Live (scaled)

Open to its target market with recurring volume.

Paused

Stopped for now, with a public reason or no recent activity.

Inactive

No longer running. The reason says why: matured or redeemed (a one-off bond or note), concluded (a pilot or trial that ended as planned), discontinued (stopped early), or superseded (replaced by a successor record). A programme stays live while any of its issues are outstanding.

3. Evidence grades

How sure we are, per source. Status and evidence are separate: a deployment can be live on B-grade evidence, or a pilot on A-grade.

A Verifiable

Regulatory filings, regulator documents, on-chain data. Something a third party can check.

B Primary, self-reported

The operator's own press release, documentation or official statement.

C Secondary

Trade or mainstream press reporting, without the primary document.

D Unconfirmed

Conference remarks, social posts, interviews. Recorded, never relied on alone.

4. Performed, supported, planned

A platform being able to do something is not the same as it having done it.

Performed

Evidence of real transactions for this use case.

Supported

Live capability, but no evidenced use yet.

Planned

Stated intent only.

When the evidence is ambiguous, the weaker label wins.

5. Last reviewed

Every record shows when it was last reviewed: the date someone re-checked it against its sources, confirmed they still say what the record claims, and looked for newer information. It tells you how recent that check is. It does not mean every claim is certain; how sure we are is the evidence grade on each source. After 90 days without a review a record turns amber; after 180, red. Stale records are re-checked, not deleted. Figures from on-chain data are refreshed on their own schedule and carry their own dates.

6. Value figures

Figures come in kinds that are never added together, because a platform's monthly turnover is not the value of the assets it holds: the balance held on the ledger now (fund assets, tokens or deposits outstanding); issued to date (bonds and tokens created since launch, including ones since matured); transacted to date (value settled, traded or paid since launch); and throughput per month (daily figures are converted using 21 business days). Each throughput or transacted figure records the use case it measures. Each figure keeps its own date and source, and new figures are added rather than overwriting old ones.

The home page cuts them three ways: by asset (tokens in circulation, or the issue size of bonds not yet matured), by platform (held now, with issued to date alongside) and by use case (throughput per month, with transacted to date alongside). Each column covers only what has been published, so the columns differ from each other; that is expected.

All totals are floors and are likely to be well below the real figures. Many live deployments publish no value at all, most of all bank and market-infrastructure services on private ledgers. Balances on public blockchains can be read directly every day, so they are counted far more completely, and this skews the totals towards stablecoins. The home page states how many live deployments report a figure, and what share of the counted value sits on public networks and in stablecoins.

Trading volume for tokenised stocks is read daily from what venues publish (exchanges' public market data and on-chain DEX pools) and is always a floor: trading in apps and on venues with no public feed isn't covered, and each figure says which venues it includes. On-chain transfer counts are shown only as background, because they include minting, redemptions and exchanges moving their own wallets. See Data health.

Where a source gives a figure for one network only, such as a fund's holdings on Solana, it is recorded against that network; this is what sizes networks on the map. Figures that describe a wider whole, such as a fund's total across all share classes, are marked as context: shown, never counted. “Up to” figures are ceilings and are left out of totals. Other currencies are converted to US dollars at ECB reference rates for totals only.

Each dollar is counted once. Tokens are often layered: a stablecoin or fund token is deposited into a vault, which issues its own token, which may be staked for yet another. We count value at the first token on a ledger that represents the real asset (the fund share, the stablecoin, the bond). A token that only holds another token we already track is marked as wrapping it: its value is shown on its own page, and its asset page says what it wraps, but it never adds to totals or the map. The home page shows a gross figure beside each headline that adds wrapper balances back, so it double counts by design. A tokenised feeder into a fund that is not itself on a ledger is the first on-ledger layer, so it counts. Where part of a token's reserve is held in other tokens we track and a dated breakdown shows how much (Sky's USDS and DAI hold USDC and tokenised Treasury funds), only the rest of its supply counts; the full supply is still shown, and added back in the gross figure. Copies of the same token on other chains, bridged by its own issuer, are one asset.

Figures are also shown as context, not counted, when the backing is mixed or not published (so part of the value may already be counted elsewhere), and while an open editorial question decides whether a kind of figure should count at all. The daily on-chain reading follows the same choice, so it never counts a figure the record itself holds back. A deployment's throughput and its total to date are always counted on the same basis, so a total to date can never look smaller than a month of activity. In every case the figure stays visible on the record, with a note saying why it is not counted.

Recorded on ledger. Some ledgers only mirror an ownership record kept elsewhere: a transfer agent's register, a central securities depository's master ledger, a lender's loan books (ledger role “record” in the classification). Where that holds for everything a deployment issues, its balances are shown and added up in a separate “recorded on ledger” figure beside the headline, not in the headline totals or the map.

Stablecoin supply, and the supply of funds whose contracts we hold, is read automatically every day from on-chain data. Readings that move suspiciously are held back for a person to check, and the last good value stays live; see Data health.

Value cases: why tokenise?

Use case pages carry our analysis of what tokenisation changes compared with how the same thing is done today. It is analysis, not reported fact, and is labelled as a draft until the editor has reviewed it. The flows are simplified for general understanding; real implementations vary a great deal and are often far more complex.

Each benefit is graded. Theory, low confidence: a credible source argues it, with little agreement or data. Theory, high confidence: widely agreed by authoritative sources but not yet measured in production. Evidenced: a live deployment in the register has a sourced outcome showing it; the site checks this automatically, and flags a benefit for review when new evidence appears. Scale is shown as marginal, meaningful or transformative, with a figure where one exists: operator-reported, from an independent study, or our estimate, which always states its assumptions. Timing says whether a benefit is available now, depends on networks growing (counterparties or cash on the ledger), or belongs to a future state.

Regulation

Regulation is recorded in three kinds. A regime is a whole framework, such as MiCA or the UK cryptoasset regime, with its key points, scope and a timeline that includes forthcoming dates. An instrument is one document within a regime: a consultation paper, policy statement, statutory instrument or rulebook chapter. A permission is a firm-specific licence, authorisation, relief or sandbox admission, with the firms that hold it.

Deployment pages show the regulation a deployment operates under, where a source says so, and separately the regimes likely relevant to it, matched on jurisdiction, asset type and activity. The second list is indicative, not legal advice.

How assets are classified

Each asset is classified on four separate questions, so no single label has to carry several meanings: its legal form (what the holder legally owns: a debt security, a fund interest, a share, a bank deposit, a stablecoin ...), its underlying (what the value comes from, looking through any fund or token it holds), its value behaviour (stable value, NAV, fixed income or market price, and whether it pays income) and its ledger role (whether the ledger is the ownership record, or a copy of one kept elsewhere). A gold fund is a fund and gold-backed, so it appears under both.

Legal form is jurisdiction-neutral and read from evidence, most authoritative first: a regulator's register, the ISO 10962 (CFI) code that comes with an ISIN, the issuer's status, then the issuance terms. Local legal labels (a MiCA e-money token, a Pfandbrief, a UCITS money market fund) are recorded separately for each jurisdiction. Where the evidence isn't published, a question is marked undisclosed rather than guessed. Every classification states its basis and confidence.

7. Controlled vocabularies

Asset types, use cases, jurisdictions, roles and network types come from fixed lists, so filters and comparisons mean the same thing everywhere. Lists are hierarchical: the European Union includes its member states, and Funds includes money market funds.

Geography works two ways. A deployment operates where it runs or which markets it serves (one deployment can have several). A firm is based where it is headquartered. Locations are mostly by headquarters: we don't record which branch or subsidiary takes part, so a London-headquartered bank's work in Hong Kong counts as a UK firm's. Regions group countries the same way across the site.

8. Data access

Every page is free to read, link to and cite. Ids are stable and never reused, so links keep working. The register as a whole is protected by copyright and database rights and is not offered as a bulk download; if you would like data for research or another use, ask us. See the terms.

Corrections

Spotted something out of date or missing? Suggest a correction, ideally with a source.

Status labels: Announced → Proof of concept → Pilot → Live (limited) → Live (scaled). See them applied in the maturity matrix.