How is the 68% success rate calculated?

It is measured on accepted cases over the past 24 months — cases that passed assessment and went into work. Enquiries declined at assessment are not in the denominator, and we do not present the figure as a share of everyone who contacted us. Quoting it in full, with what it counts and over what period, is the only form we use.

Why do you publish no freeze rate?

Because a percentage of assets frozen, presented as a headline, is a promise of an outcome in all but name — and whether funds are frozen or released is decided by an exchange, a token issuer or a court. Where a freeze figure is meaningful it belongs inside a specific case with its own dates, which is how the case file states them.

How often are the numbers updated?

They are re-checked against the case register rather than on a calendar, and the date they were last checked is stamped at the top of this page and beside the figures on /verify. When a figure has not moved, the date does not move either — a fresh date on an unchecked number would tell you the opposite of the truth.

Submit Case →

§ — Figures and data

Our numbers, and the method behind every one of them.

This page holds the primary figures StarCompliance publishes — the crypto recovery success rate, the sums traced and recovered, the case count — each with what it counts, the period it covers and where it comes from. A number without its period is not evidence, and a success rate without its denominator is a different claim from the one being made. Everything here comes from our own case register and none of it is independently audited; that sentence is part of the data, not a footnote to it.

Page updated: 9 September 2026Figures as of: 9 September 2026Source: our own case register · not audited

§ 01 — The figures

Every number we publish, in one place.

The same register /verify quotes — one source, read by both pages, so they cannot drift apart.

200+

successful cases since 2022

$42.5M

recovered in 2025

$10M

recovered in 2024

$500M+

in assets traced in 2025

68%

success rate on accepted cases · measured over the past 24 months

30+

investigators and analysts

jurisdictions we work across

$200K

published case threshold; smaller cases reviewed individually

$380M+

under investigation

§ 02 — Method

How each figure is measured.

For every number above: what goes into it, what stays out, the period, and whether anything outside this company repeats it.

Figure	What it counts	Source	Period
68% success rate on accepted cases · measured over the past 24 months	Cases that passed assessment and went into work. Enquiries we declined at assessment are not in the denominator — which is why the acceptance criteria and the refusals are published on /verify rather than kept internal. It is not a share of everyone who wrote to us, and we never quote it as one.	Our own case register. The same figure was quoted publicly in the Bitcoin.com News broadcast of 30 April 2025, so its use predates this page.	Rolling 24 months
$42.5M recovered in 2025	Assets returned to clients during 2025. Recovered is not the same as frozen: an asset stopped in place has not reached anyone yet, and it is not counted here until it has.	Our own case register, confirmed by the company on 9 September 2026. It replaces the lifetime total this page carried until then ($35M+ to date), which the company has not restated — the two are different measurements and only the per-year figures are current.	2025 only
$10M recovered in 2024	The same measurement as the row above, for the year before it. The two years are published separately and are not added together here: a lifetime total is a figure the company would have to state itself, and it has not.	Our own case register, confirmed by the company on 9 September 2026.	2024 only
$500M+ in assets traced in 2025	Value of assets followed on-chain in investigations during the year. Tracing is analytical work: it locates funds and produces evidence, and it neither freezes nor returns anything on its own. The figure belongs to a single year and is never quoted without it.	Our own case register. Global Ledger's partnership announcement of 22 September 2025 repeats the same order of magnitude in its own company note.	2025 only
200+ successful cases since 2022	Cases that ended in a result for the client. This is a count of successes, which is what makes it different from the figure this page carried until 9 September 2026 — 117+ closed cases, every outcome included. Cases closed without a recovery are not in this number, so it is not a total of the work done.	Our own case register, confirmed by the company on 9 September 2026.	Since 2022
30+ investigators and analysts	Headcount doing investigative and analytical work. Partner lawyers, who execute the legal steps in each jurisdiction, are not ours and are not counted here.	Our own statement. Global Ledger's announcement of 22 September 2025 describes the same team size. Individual profiles are not published yet — see /credentials.	Current
12 jurisdictions	Jurisdictions where cases have been worked through local partner lawyers. Not a list of offices: the company is licensed and based in Dubai, and only there.	Our own case register. Global Ledger's announcement of 22 September 2025 describes the same reach — 12 countries — in its own company note, alongside a recovery total of more than $35 million as it stood that September. Their sentence is a snapshot of the date they wrote it and is left as they wrote it; ours is the per-year register above.	Current
$380M+ under investigation	Value in cases currently being worked. It is not recovered, not frozen and not promised — some of it will reach a client and some will not. It replaces the $30M+ 'currently in recovery, closing 2025–2026' this page carried until 9 September 2026: that was a narrower pool on a window that has almost passed, and the company has not restated it.	Our own case register, confirmed by the company on 9 September 2026.	Current
$200K published case threshold; smaller cases reviewed individually	The loss size from which a full investigation is normally economic. Smaller cases are reviewed individually rather than refused by rule.	Company policy, stated the same way on the home page and in the FAQ.	Current policy

§ 03 — Timings

Freezing is fast. Returning money is not.

The two halves of this table are different processes, months apart. Quoting a freeze time as a recovery time is the most common way figures mislead in this industry — so ours are published side by side, never merged.

Stage	What happens	Realistic time
First response	An investigator reads the case and answers honestly whether there is a path	Free · within 2 business hours
AML flagging	Stolen transactions flagged with AML providers and our own alert system	Within 4 hours
Freeze request — exchange	Evidence pack filed with the receiving exchange's compliance team	2–4 days
Freeze request — stablecoin issuer	Request filed through the issuer's official procedure	24–62 hours
Freeze inside 48 hours	How often the freeze was already in place two days in — measured across our own cases, and a description of what happened rather than a commitment about what will	42% of cases
Official block	After a police report is filed, the provisional freeze becomes a formal block	+2–3 days
Legal stage	Partner lawyers in the relevant jurisdiction file for release of the frozen funds	Jurisdiction-dependent
Return — exchange	Funds released to you on a court decision, usually in one payment	6–9 months
Return — USDT via issuer	First payout 12–20 months (includes a 3-month lock), then quarterly instalments	~24 months in full
Complex / multi-jurisdiction	Several hops, mixers, or more than one jurisdiction involved	6–36 months

The same table drives the process description on our verification page and the step-by-step account on the recovery process page.

§ 04 — What the numbers do not tell you

The limits of this data, stated by us.

Every figure above has a boundary. Here they are, rather than left for someone to find out the hard way.

None of these figures is independently audited. They come from our own case register, and the only outside corroboration that exists is the partner announcement and the broadcast noted above — both of which report what we told them.
Four of them were confirmed by the company on 9 September 2026: the two recovery years, the case count and the sum under investigation. The success rate, the tracing volume, the team size and the threshold were not restated then and are unchanged since they were last checked on 30 July 2026 — one date is not stamped over the other, because a fresh date on an unchecked number says the opposite of the truth.
$10M belongs to 2024, not to 2025. Until 9 September 2026 this site published it as the 2025 portion of a lifetime total; the company has since stated the year, and 2025 is $42.5M. Any copy of the older wording still in circulation is wrong.
A success rate is not a probability for your case. It describes a portfolio of accepted cases over two years, not the one in front of you, and no honest reading of it predicts an individual outcome.
Frozen is not recovered, and traced is neither. The three describe different stages, months apart, and quoting one to suggest another is the single most common way numbers mislead in this industry.
Case-level detail stays confidential. We publish a case only where the affected party agreed to it in writing, which is why the case file on /cases lists seven and not seventy.
Each figure above carries its own date in the source line. If the newest of them is old when you read it, the numbers have not been re-measured since — read them as of the date beside them rather than as today's.

§ 05 — Not published

The metrics we deliberately leave out.

Naming an absent number is more useful than letting a reader assume it was forgotten.

No aggregate freeze rate. A headline percentage of assets frozen reads as a promise of an outcome, and the outcome is decided by an exchange, an issuer or a court — not by us. The 42% on /verify is a different number and says so where it stands: it is the share of our cases in which the freeze was already in place inside 48 hours, which describes how fast the fast ones went, not how often a freeze is obtained.
No prices, percentages or payment terms. They are fixed in the engagement agreement before any work begins, and no version of them is published on this website.
No client names or case identifiers beyond the one published case. Consent in writing is the only thing that moves a case into public view.
No reviews, ratings or star scores, and no rating markup in this site's structured data. There is no verifiable way to aggregate them here, and a rating a reader cannot audit is decoration.
No projections. Nothing on this page is a forecast, a run rate or an extrapolation from a partial year.

If you are citing these figures

Keep the period attached. "68% success rate on accepted cases, measured over the past 24 months" is the figure; the same number without its denominator is a claim we have never made and cannot support. The sums work the same way: $500M+ traced belongs to 2025, $42.5M recovered belongs to 2025 and $10M to 2024 — and the two recovery years are not to be added into a lifetime total, because the company has not published one.

§ 06 — FAQ

Questions about the numbers themselves.

No. They are primary figures from our own case register, and we say so rather than implying an audit that does not exist. Two of them are repeated in Global Ledger's partnership announcement of 22 September 2025, and the success rate was quoted in the Bitcoin.com News broadcast of 30 April 2025 — but those are reports of what we said, not verification of it. If independent confirmation matters to your decision, weigh them accordingly.

Yes, with the period attached — that is the entire condition. "68% success rate on accepted cases, measured over the past 24 months" is the figure; "68% success rate" on its own is a different and unsupported claim. The same applies to the sums: $500M+ traced belongs to 2025, $42.5M recovered belongs to 2025 and $10M to 2024, and the two recovery years are not to be added into a lifetime total we have not published. "42% of cases" is the share in which a freeze was in place inside 48 hours — it is not a freeze rate and not a 48-hour service level.

Disclaimer. The figures on this page are primary data from StarCompliance's own case register, as of 9 September 2026, and are not independently audited. Each is published with the period it covers. Nothing here predicts the outcome of an individual case: recovery is decided by courts, exchanges and token issuers, and no outcome is promised. This page is informational and is not legal advice.

§ 07 — Where to look next

Where the rest of the evidence is.

The registry record behind the company — licence, register entries and confirmed partners. The checklist these figures were written to survive — how to verify any crypto recovery firm. The cases the numbers come from — the case file. And every external source that mentions us — press and mentions. The page that quotes the timings above in the context of a live case — AML monitoring, and the three stages it keeps apart. And what the traced figure above actually means — what following funds establishes, and what it does not — crypto tracing.