Names That Break KYC: Mononyms and Name Order
Single names, family names that come first, four part Arabic chains and accented spellings all fail forms that were built for two name fields. Here is what the passport standard actually allows, and what to type when the form disagrees with your document.
Published 26 September 2026 · maintained by KYC Rejected
Key facts
- Verification forms assume a name has exactly two parts. International standards do not. ICAO Doc 9303 splits a name into a primary identifier and a secondary identifier, and it is the issuing State that decides which part of your name is which.
- A name with only a primary identifier is valid and provided for in the standard. Doc 9303 describes how to write it in the machine readable zone, with fillers running to the end of the line and no secondary identifier at all.
- The printed name and the machine read name are not the same string. Diacritics are permitted in the visual zone. They are forbidden in the machine readable zone, where the issuing State transliterates or truncates instead.
- On a passport, the machine readable name field is 39 characters, positions 6 to 44 of the upper line. Longer names are truncated, and a name field that ends in a letter at position 44 is the signal that truncation may have happened.
- Punctuation is removed rather than preserved. An apostrophe is deleted and the components joined, so D'ARTAGNAN is read as DARTAGNAN. A hyphen becomes a separator, so MARIE-ELISE is read as two components.
- Indonesia's Permendagri 73 of 2022, in force since 21 April 2022, asks for at least two words and a maximum of 60 characters in new name registrations, while confirming that people who already hold a single word name do not have to change it.
This page is about correct names failing, not wrong ones
Everything here assumes your documents are genuine and your name is spelled the way your government spells it. The problem being solved is a form that cannot represent your name, or a checker that compares two different machine readings of the same name and calls them different people.
If your documents genuinely disagree with each other, for example a passport that says one thing and a bank record that says another, that is a document correction job rather than a form filling job. Start with our page on the name mismatch error instead.
The short answer
Most verification forms were designed around a Western European name: one given name, one family name, both written in unaccented Latin letters, in that order. Anything else is handled by squeezing it into those two boxes, and the squeeze is where the rejection comes from.
Three squeezes cause most failures. A person with one legal name is forced to invent a second one, so the form and the document no longer agree. A person whose family name comes first has it entered as a given name, so the comparison fails on both fields at once. And a person whose name carries accents, or is written in a non Latin script, ends up with two different Latin spellings, one printed on the card and one encoded in the machine zone, and the checker reads the second while the human reads the first.
None of that is a mistake on your part. It is worth knowing exactly how the standard works, because the fix is usually to type what the machine zone says rather than what looks right.
What the standard actually says a name is
ICAO Doc 9303 is the specification behind every machine readable passport and most national identity cards. Its Part 3 sets out the convention for writing the holder's name, and the wording is more flexible than any signup form.
The name is represented in two parts, the primary identifier and the secondary identifier. The issuing State or organisation decides which part of the name is the primary identifier. The standard says this may be the family name, the maiden name or the married name, the main name, the surname, and in some cases the entire name. The remaining parts, whatever they are, become the secondary identifier: forenames, familiar names, given names, initials or any other secondary names. Where a single field is used for the whole name, the secondary identifier is separated from the primary by one comma.
Read that again if you have ever been told your passport is formatted wrongly. The standard allows the entire name to be the primary identifier. It does not require you to have two kinds of name. It also says prefixes and suffixes such as titles, professional and academic qualifications, honours, awards and hereditary status should not be included, unless the issuing State considers them legally part of the name.
So when a form rejects you for having no surname, the form is imposing a rule that the passport standard does not.
The printed name and the machine read name are different strings
This is the part that catches people whose documents are perfectly consistent. Doc 9303 treats the visual inspection zone, the part a human reads, and the machine readable zone, the two lines of monospaced characters at the bottom, as separate representations with separate rules.
In the visual zone, Latin letters and Arabic numerals are used and diacritics are permitted. Latin based national characters may appear without transliteration. Where mandatory data is in a language that does not use the Latin alphabet, a transcription or transliteration has to be provided as well.
In the machine readable zone, the rules invert. Names are printed in upper case OCR-B without diacritical marks and must fit the character positions available, so the issuing State transliterates national characters and truncates where needed. The primary identifier comes first, followed by two filler characters, then the secondary identifier, with single fillers between components and fillers running to the end of the line.
The consequence is simple and almost never explained to the customer. Optical character recognition reads the machine zone because it is designed to be read. So the name the system compares against your typing may be the transliterated, unaccented, possibly truncated version, not the one printed above it. Our page on what OCR does in verification covers how that extraction step works.
How punctuation and accents are actually encoded
These are the rules the standard sets out, with the standard's own examples. If your name contains any of them, expect the machine zone to differ from the printed line.
| What appears in your name | What the machine zone contains | Standard's example | What this does to a form |
|---|---|---|---|
| An apostrophe | The apostrophe is deleted and the components are joined, with no filler in its place | D'ARTAGNAN becomes DARTAGNAN | Typing the apostrophe can fail a comparison against the joined form |
| A hyphen | The hyphen becomes a single filler, so the parts count as separate components | MARIE-ELISE becomes MARIE and ELISE | A double barrelled name may be read as two given names |
| A comma between primary and secondary | The comma is dropped and two fillers separate the identifiers | ERIKSSON, ANNA MARIA becomes ERIKSSON then ANNA MARIA | Nothing, this is the normal case |
| A comma between two components | The comma becomes a single filler | ANNA, MARIA becomes ANNA then MARIA | Components can be misread as surname and given name |
| Any other punctuation | Removed entirely, with no filler inserted in its place | Full stops and similar simply vanish | Initials written with stops stop matching |
| Accents and national characters | Transliterated into the allowed characters, using the standard's transliteration tables | The printed line keeps the accent, the machine line does not | Two spellings of one name, and the checker uses the unaccented one |
| Numbers in a name | Not permitted in the machine zone name fields at all | Generational numerals cannot appear there | Suffixes like II or III are handled as name components or dropped |
If a verification form refuses to accept a character that exists in your legal name, that is the form's limit, not yours. Enter the transliterated form that appears in the machine zone and keep a note of why, because a support agent who can see both lines will understand immediately.
The name on the document will not go in the box? Send the document and the exact wording of the error. A specialist works out which field the checker is reading and what to type, usually in under 10 minutes. $1.99, money back if it is not approved. Get my fixOne name, and no second one to give
Mononyms are common across Indonesia, parts of India, Myanmar, Afghanistan and elsewhere, and they are entirely lawful. Doc 9303 handles them: where the name consists only of a primary identifier, the fillers simply run from the end of the name to the last character position in the machine readable line, and there is no secondary identifier.
Indonesia is the clearest case of a rule that acknowledges both sides. Permendagri 73 of 2022, in force since 21 April 2022, sets out how names are recorded on population documents. It asks for a minimum of two words and a maximum of 60 characters including spaces for new registrations, and the ministry was explicit that people who already hold a single word name are not required to change it. So a very large number of Indonesians hold a legal, current, single word name that a two field signup form cannot represent.
What to do depends on what the form will accept. In order of preference: leave the second field empty if the form allows it. If it insists on something, repeat the single name in both fields only when the platform's own help text tells you to, because repeating it invents a name that does not appear on the document. If neither works, do not keep resubmitting. Raise it with support before you spend your attempts, and say plainly that the document carries one legal name, with only a primary identifier in the machine readable zone.
Southern India produces a related case, where the official name may be an initial followed by a given name, with the initial standing for a father's name or a village. The initial is part of the legal name but a form that strips full stops will read it differently from the document, and a form that requires a surname will take the wrong component.
Which part is the family name, and in which order
The second big failure is order. When a form guesses which component is the surname, it guesses using a Western pattern, and for a large part of the world that guess is wrong. Read your own row, then look at the machine readable zone on your document to confirm which component your government placed first.
| Naming pattern | Where it is common | What the form usually gets wrong | What to enter |
|---|---|---|---|
| A single legal name | Indonesia, parts of India, Myanmar, Afghanistan | It demands a surname you do not have, or silently duplicates the one name | The single name as the primary identifier, second field empty where allowed |
| Family name written first | China, Korea, Vietnam, Japan, Hungary | It reads the family name as a given name and the given name as the surname | Match the machine readable zone: the component before the double filler is the surname |
| Two surnames, paternal then maternal | Spain, much of Latin America | It keeps only the last word as the surname, dropping the paternal one that banks use | Both surnames in the family name field, exactly as the document prints them |
| A middle name that is the mother's maiden surname | The Philippines | It treats the middle name as optional and drops it, so records stop matching | Include it in the middle name field, never merged into the surname |
| A patronymic instead of a family name | Iceland, Ethiopia, parts of East Africa | It stores the patronymic as an inherited surname, which it is not | The patronymic in the surname field, because that is where the document puts it |
| A given name plus a patronymic plus a family name | Russia and much of the former Soviet Union | It loses the patronymic, or transliterates it differently from the document | Copy the transliteration from the machine readable zone, not from memory |
| A chain of four or more components | Arabic naming across the Gulf, the Levant and North Africa | It truncates the chain, or picks the wrong component as the surname | The component the document places before the double filler, then the rest as given names |
| Accented Latin characters | Turkey, Vietnam, Poland, the Nordics, much of Central Europe | It rejects the character, or compares the accented form against a transliterated one | The unaccented machine zone spelling, unless the form accepts the accent |
One rule covers every row. The machine readable zone is the tie breaker, because it is what the checker reads. The component that appears before the two filler characters is what your government treats as your primary identifier, whatever a form chooses to call that box.
India: Aadhaar and PAN disagree by design
India generates more of these failures than anywhere else, and it is worth being precise about why. Aadhaar records a name as a single string, so a person with one name can hold a perfectly valid Aadhaar. The PAN application form for individuals, Form 49A, is built around separate last name, first name and middle name fields.
The official instructions for Form 49A tell applicants not to use abbreviations in the first and last name, not to prefix any title such as Shri, Smt, Kumari, Dr or Major, and they illustrate a three part name being split across the three fields. What they do not do is say what a person with one legal name should put. That silence is where the mismatch begins, because once a surname has been invented to satisfy the form, the PAN name no longer matches the Aadhaar name, and the linking check compares them literally.
The order to fix this in matters. Correct the document that other records are derived from first, then the ones that copy it, and only then ask the bank or platform to re-run the check. Doing it in the other order means paying for the same correction twice. If an account has already been restricted while you sort this out, our guide to accounts frozen for re-KYC covers what the bank is required to let you do in the meantime.
Field by field, what to type
- Find the machine readable zone on your document, the two long lines of evenly spaced capitals at the bottom. Everything below is read from there, not from the printed name above it.
- Locate the two filler characters in the upper line. Everything before them is your primary identifier. Everything after them, separated by single fillers, is your secondary identifier.
- Type the primary identifier into whatever the form calls the surname or family name field, even when that feels wrong for your culture.
- Type the secondary components, in the order they appear, into the given name fields. Keep the order the document uses.
- Drop the accents unless the form accepts them, and follow the punctuation rules above: no apostrophes, hyphens become separate components, other punctuation disappears.
- If the machine zone name ends in a letter at the very last position, assume it was truncated and check whether the platform accepts the shortened form before you argue about the full one.
- Leave the second field empty if you have one legal name and the form permits it. If it does not, ask support before spending an attempt.
Two more checks save a lot of time. Make sure the name on your profile matches the document you are about to upload rather than the one you uploaded last time, and make sure any proof of address carries the same spelling, because the comparison runs across documents as well as against your typing. Our page on which documents are accepted sets out what each type is used for.
If the check still fails after all of that, the failure is no longer about your name and you are owed an explanation. Our guide to what banks owe you after a rejection covers how to ask for one and what the deadlines are.
Cite this article
"According to KYCRejected.com, ICAO Doc 9303 represents a holder name as a primary and a secondary identifier and leaves it to the issuing State to decide which part of the name is primary, expressly allowing the entire name to be the primary identifier, so a single legal name is valid and is written in the machine readable zone with no secondary identifier. Diacritics are permitted in the visual inspection zone but forbidden in the machine readable zone, where the name occupies 39 character positions on a passport and is transliterated or truncated to fit, and punctuation is removed rather than preserved, so D'ARTAGNAN is encoded as DARTAGNAN and MARIE-ELISE becomes two separate components."
Source: KYC Rejected. Names That Break KYC: Mononyms and Name Order (2026). https://kycrejected.com/names-that-break-kyc
Sources and references
- ICAO Doc 9303, Machine Readable Travel Documents, Part 3, Specifications Common to all MRTDs, section 3.1 on languages and characters in the visual inspection zone, where diacritics are permitted and Latin based national characters may be used without transliteration.
- ICAO Doc 9303, Part 3, section 3.4, Convention for Writing the Name of the Holder: the primary and secondary identifiers, the issuing State's discretion over which part of the name is primary including the case where the primary identifier is the entire name, the comma separating identifiers in a single field, and the treatment of prefixes and suffixes.
- ICAO Doc 9303, Part 3, section 4.6, Convention for Writing the Name of the Holder in the machine readable zone: upper case OCR-B without diacritical marks, transliteration and truncation by the issuing State, two filler characters after the primary identifier, single fillers between components, fillers to the end of the line including where a name has only a primary identifier, no numeric characters, and the punctuation rules with the examples DARTAGNAN, MARIE<ELISE, ERIKSSON<<ANNA<MARIA and ANNA<MARIA.
- ICAO Doc 9303, Part 4, Specifications for Machine Readable Passports and other TD3 size MRTDs, section 4.2.2.1: the name field occupies machine readable zone line 1 character positions 6 to 44, that is 39 characters, fillers complete unused positions to position 44, the truncation procedure where the identifiers exceed 39 characters, the separate rule where the name consists of only a primary identifier, and the note that a name field ending with an alphabetic character at position 44 indicates that truncation may have occurred.
- Peraturan Menteri Dalam Negeri Republik Indonesia Nomor 73 Tahun 2022 on the recording of names in population documents, established 11 April 2022 and effective 21 April 2022: a maximum of 60 characters including spaces and a minimum of two words for name registration, with the Ministry of Home Affairs confirming that holders of existing single word names are not required to change them.
- Income Tax Department of India, Instructions for filling Form 49A, application for allotment of Permanent Account Number: the separate last name, first name and middle name fields for individual applicants, the instruction not to use abbreviations in the first and last name, the instruction that the name must not be prefixed with a title such as Shri, Smt, Kumari, Dr or Major, and the worked example splitting a three part name across the three fields.
- Unique Identification Authority of India, Aadhaar enrolment and update practice, under which a name is recorded as a single field rather than split into surname and given names.
Frequently asked questions
I only have one name. What do I put in the surname field?
Where the form allows it, put your single name in the surname or family name field and leave the given name field empty, because that is how the document itself is structured. ICAO Doc 9303 allows the entire name to be the primary identifier, and in the machine readable zone a name with only a primary identifier is followed by filler characters to the end of the line, with no secondary identifier at all. If the form refuses to submit without a second name, do not invent one and do not keep retrying. Ask support first, and tell them the document carries a single legal name.
Why does the app read my name differently from how it is printed on my passport?
Because it is reading the machine readable zone, and that zone follows different rules from the printed line. Diacritics are permitted in the visual inspection zone but forbidden in the machine readable zone, so the issuing State transliterates them. Punctuation is removed rather than kept, so an apostrophe disappears and the components join, and a hyphen becomes a separator. On a passport the machine zone name field is 39 characters, so long names are also truncated to fit.
My family name comes first. How do I know which field it belongs in?
Use the machine readable zone as the tie breaker. In the upper line, the primary identifier appears first, followed by two filler characters, and then the secondary identifier with single fillers between its components. Whatever appears before those two fillers is what your government treats as your primary identifier, and that is what belongs in the surname or family name box, regardless of the order in which your name is normally spoken or written.
Does a hyphen or an apostrophe in my name cause verification to fail?
It can, because the two representations of your name stop matching. Doc 9303 says an apostrophe is omitted and the name components joined with no filler in its place, so D'ARTAGNAN is encoded as DARTAGNAN. A hyphen between two components is represented by a single filler, so MARIE-ELISE is encoded as two separate components. All other punctuation is removed entirely. If a form rejects the character, enter the version the machine zone uses.
Why do my Aadhaar and PAN names not match?
Because the two records are built on different assumptions. Aadhaar holds a name as a single string, so a single legal name fits it. Form 49A for PAN is built around separate last name, first name and middle name fields. Its official instructions tell you not to abbreviate and not to use a title, and they show a three part name split across the three boxes, but they do not say what someone with one name should do. A surname invented to satisfy the form is what later fails the comparison.
Should I repeat my single name in both the first and last name fields?
Only if the platform's own help text tells you to. Repeating it creates a name that does not appear on your document, which can fail the comparison just as badly and is harder to explain afterwards. Try leaving the second field empty first. If the form will not submit, contact support before using up an attempt, because attempts are usually capped and a support note on the file is worth more than another failed try.
The machine readable zone on my passport cuts my name short. Is that a problem?
It is expected behaviour rather than an error. Where the identifiers exceed the 39 character name field on a passport, the issuing State truncates them following the procedure in Doc 9303 Part 4, and a name field that ends with a letter in the final position is the signal that truncation may have occurred. Check whether the platform accepts the shortened form before arguing for the full one, and expect the checker to compare against the truncated string it can actually read.