Names rarely match exactly.
The same person is Müller on one list, Mueller on another and Muller on a third. Character-by-character comparison misses all three. Krino flattens the differences first, then compares.
After the name, not the spelling.
Accents and special letters
Characters like é ü ø ß æ ı ğ ş are reduced to their base forms. Cases that standard decomposition does not solve — Turkish dotless ı among them — are handled explicitly.
Name order and extras
Given-name order, middle names, titles and punctuation differences do not break a match. The comparison works over name parts.
Similarity threshold
Close-but-not-identical spellings are scored by similarity. You set the threshold — you are the only party who knows how many false alarms you can absorb.
What gets screened?
Sanctions lists
Consolidated lists from the major sanctions regimes are gathered into one source and refreshed regularly. People and organisations go through the same screening.
Politically exposed persons
People who require heightened scrutiny because of public office are flagged as their own category. A PEP match is usually a reason to look, not a reason to decline — Krino lets you keep the two apart.
Your own watchlists
Internal deny lists, people you have previously caught, or a list shared by a partner — screened with the same name matching, which is to say with the same tolerance as the published lists.
The difference matters: a list of names that demands exact equality never fires in practice. A name reaches you spelled however the customer's system stored it, and only the entries somebody typed identically in both places are caught. The list looks full and does nothing.
Scheduled re-screening
Lists change; a customer who was clear yesterday may be listed today. Krino re-screens your existing customer base at an interval you choose and opens cases for new matches.
An address is the opposite of a name.
Screening now covers crypto wallet addresses alongside names — but not on the same logic. A name has many spellings; an address has exactly one form and exactly one owner.
Where the list comes from
OFAC's SDN list. As a work of the US federal government it is in the public domain and carries no licensing restriction — which is why it is the one that ships.
Roughly a thousand addresses across about a hundred designated parties, spanning around twenty asset codes. Bitcoin and TRON dominate — which is what illicit flows actually use.
Exact, not similar
Unlike name screening, address comparison is character for character, and deliberately so. Something close to a listed address is not a weaker hit; it is somebody else's address.
One canonical form per chain
The comparison runs on a canonical form chosen per chain. An Ethereum-style 0x address folds to lower case; a Bitcoin or TRON base58 address keeps its case exactly — lower-casing one of those produces a string that is not a valid address at all.
Refreshed daily, never silently broken
The list is refreshed every day and reaches a running host without a restart. A download that arrives empty, too small or unparseable is refused rather than applied, and the list already in use stays in use.
A publish can break at the other end. What must not follow from that is screening quietly switching itself off.
If the list never loaded on that host, the address is treated as a match. Calling an address that could not be screened "not listed" is not a weakened control; it is a clearance nobody performed.
Not every name arrives with a transaction.
The engine screens every party a rule points it at. But most of the names a compliance team wants to check never become a transaction: a customer being onboarded, a payee checked before releasing a held file, a name that arrived on a phone call.
One name, one screen
Type the name and set the threshold where you want it. Scored matches, each row showing the listed party, which spelling matched, its type, its countries and which list it came from.
The engine itself
Not a separate search: the same list and the same matcher the rule path uses. What this screen says about a name is what the engine would say about it.
Who, what, when
Every search is recorded. Screening by hand is a compliance act that leaves no other trace — no decision, no file. It is also exactly what an audit asks for.
A server whose list has not loaded refuses the search rather than saying "no match". They are different sentences, and in sanctions screening the difference is the one between a control and a formality.
A match is a question, not a decision.
A name matching does not automatically decline the transaction. The match enters your rules as a signal; what happens next is up to you.
A match is found
Which list, which record and at what similarity — all recorded.
Rules weigh it
A sanctions match declines outright, a PEP match opens a review — as you configured it.
A case opens
Anything needing review lands in the team's queue with the matched record inside the file.
The outcome is recorded
False match or real — who decided, when and how, in the audit trail.
Let's screen your current book.
How many matches does today's customer list produce, and how many are false alarms? Let's find out together.