UK Postcode Format Explained: Outward and Inward Codes

If you’ve ever typed a postcode into a checkout form and had it rejected, or wondered why “SW1A 1AA” doesn’t look anything like “M1 1AE,” you’ve run into the quirks of the UK postcode format. Unlike the fixed-length numeric codes used in many countries, UK postcodes are alphanumeric, variable in length, and built from two distinct parts that each do a different job.

This guide breaks down exactly how a UK postcode is put together — the outward code, the inward code, the space between them, and what each character is actually telling a sorting machine or a delivery driver. It also covers where format checks stop and real postcode validation begins, since that distinction trips up a lot of people building address forms.

What Is a UK Postcode?

A UK postcode is a short alphanumeric code, created and maintained by Royal Mail, used to route mail to a specific group of addresses or a single large delivery point. It’s typically five to seven characters long (not counting the space) and combines letters and numbers in a way that narrows down location progressively — from a broad postal area down to a small handful of addresses.

Postcodes were rolled out nationally between 1959 and 1974, originally to support mechanised mail sorting. Today they’re used well beyond the postal system — in insurance pricing, sat-nav routing, delivery checkout forms, census data, and location-based services generally.

A full postcode does not usually point to one single building. In most cases it identifies a small cluster of roughly ten addresses, or one major point such as a large office, hospital, or business that receives enough mail to warrant its own code.

The Two Parts of Every UK Postcode: Outward Code + Inward Code

Every standard UK postcode splits into two halves, separated by a single space:

Outward code + Inward code

Take SW1A 1AA as an example:

SW1A is the outward code — it gets the item to the right sorting office or delivery area.
1AA is the inward code — it narrows that down to a specific street, building, or small group of addresses within that area.

The outward code varies in length (two to four characters). The inward code, by contrast, is always exactly three characters: one digit followed by two letters. That fixed length for the inward code is one of the most useful things to know if you’re validating postcodes programmatically, since it’s the one part of the format that never changes.

The Outward Code Explained

The outward code is the first block of characters, before the space. It generally identifies:

The postcode area — one or two letters representing a broad region or city, usually derived from a major post town (for example, SW relates to South-West London, EH to Edinburgh, M to Manchester).
The postcode district — one or two digits (occasionally followed by a letter) that narrow the area down to a specific district within it.

Outward codes aren’t a fixed length because postal areas and districts themselves vary in size and mail volume. A short outward code like M1 covers a large amount of mail in a compact way, while longer ones like SW1A exist mainly in busy areas — particularly parts of London — where an extra letter is needed to distinguish neighbouring districts.

A small number of outward codes are non-geographic. These are assigned to specific large organisations rather than a location, so they don’t map to a physical area in the way most outward codes do.

It’s worth being careful not to oversimplify: an outward code identifies an area or a district, not a single address. Two properties on opposite sides of a district can share the same outward code while having completely different inward codes.

The Inward Code Explained

The inward code is the second block of characters, always appearing immediately after the space, and it is always exactly three characters long: one number followed by two letters.

The number identifies the postcode sector — a subdivision of the district. Sectors are typically numbered 1 through 9, then 0 (historically treated as the tenth sector rather than the first, a quirk left over from how Royal Mail originally ordered them).
The two letters identify the postcode unit — a small group of addresses, sometimes as few as one, within that sector.

Because the inward code’s length and structure are fixed, it’s genuinely useful for narrowing delivery down to street level or a specific building, which is why sorting offices rely on it heavily during the final stage of routing mail. A handful of letters (C, I, K, M, O, V) are deliberately avoided in postcodes generally, largely because they’re easy to confuse with digits or other letters when handwritten.

UK Postcode Format Patterns

Not every postcode follows an identical shape. Royal Mail’s system allows for six main structural patterns, all built from the same principle — letters for the area, digits for the district, then a space, then a digit and two letters for the inward code.

Pattern	Example	Structure
A9 9AA	M1 1AE	1 letter + 1 digit, space, 1 digit + 2 letters
A99 9AA	M60 1NW	1 letter + 2 digits, space, 1 digit + 2 letters
AA9 9AA	CR2 6XH	2 letters + 1 digit, space, 1 digit + 2 letters
AA99 9AA	DN55 1PT	2 letters + 2 digits, space, 1 digit + 2 letters
A9A 9AA	W1A 0AX	1 letter + 1 digit + 1 letter, space, 1 digit + 2 letters
AA9A 9AA	EC1A 1BB	2 letters + 1 digit + 1 letter, space, 1 digit + 2 letters

A couple of things worth flagging: these are generic format patterns, not confirmed real postcodes — matching one of these shapes tells you a string is structured correctly, not that it’s an actual assigned postcode. W1A 0AX, above, happens to be a well-known real example (linked to the BBC’s central London address), but generic examples elsewhere in this guide are for illustration only.

Where the Space Goes — and Why It Matters

The single space sits between the outward code and the inward code, and it isn’t optional in properly formatted output — for example:

SW1A 1AA

In practice, people write postcodes inconsistently, and most systems need to handle that:

Missing the space (SW1A1AA) — still identifiable, since the inward code’s fixed three-character length lets software work backward from the end of the string to reinsert the space correctly.
Space in the wrong position (SW1 A1AA) — this breaks that logic and typically causes a lookup or validation failure.
Extra or inconsistent spacing (SW1A 1AA or SW 1A 1AA) — usually needs to be normalised (spaces stripped, then reinserted before the final three characters) before the postcode can be matched against a database.

A common, reliable approach for handling free-text input is: strip all spaces and hyphens, uppercase the string, then insert a single space before the last three characters. That’s formatting normalisation, though — it doesn’t confirm the postcode is real (more on that below).

What the Letters and Numbers Represent

Putting the outward and inward code together, a UK postcode encodes four levels of geography, each nested inside the one before it:

Postcode area — the leading letters (e.g. the SW in SW1A 1AA), covering a broad region.
Postcode district — the area plus the following digits (e.g. SW1A), narrowing to a specific district.
Postcode sector — the district plus the first character of the inward code (e.g. SW1A 1), narrowing further.
Postcode unit — the full postcode, right down to the final two letters, identifying a small group of addresses or a single large delivery point.

This is a hierarchy, but it isn’t perfectly uniform across the country — some areas are subdivided more finely than others depending on population density and mail volume, so it’s inaccurate to assume every character plays an identical geographic role in every postcode.

UK Postcode Examples by Structure

A few illustrative examples, showing how the same underlying logic produces different-looking postcodes:

One-letter outward area: M1 1AE — the outward code is just a single letter and single digit.
Two-letter outward area: CR2 6XH — two letters identify the area before the district digit.
Letter–number–letter outward code: W1A 0AX — a trailing letter after the district number, common in busy central London districts.
Longer outward code: DN55 1PT — two letters plus two digits, used where a district number needs two digits to stay unique.

In each case, the inward code keeps the same shape: one digit, two letters, always three characters total.

UK Postcode vs Full Address

A postcode is not a substitute for a complete address, and it’s worth being explicit about the difference:

A postcode typically groups around ten addresses, or represents one high-volume delivery point — it is not, by itself, a single unique building in most cases.
Multiple properties on the same street, or in the same block of flats, often share one postcode.
A complete UK address normally needs a building number or name, street name, post town, and the postcode together — the postcode narrows the delivery area, but the earlier lines of the address do the rest of the work.

This matters for anyone designing address-lookup forms: a postcode-only search will usually return a short list of properties to choose from, not one definitive result.

Format Validity vs Real Postcode Validity

This is one of the most commonly confused points, so it’s worth separating clearly:

Format checking confirms that a string matches one of the six valid shapes — right number of letters and digits, in the right order, with a space in a plausible place. This can be done with a simple pattern check.
Real-postcode validation confirms the postcode is actually assigned and in use — which requires checking it against Royal Mail’s Postcode Address File (PAF) or an equivalent up-to-date postcode database, not just a formatting rule.

A string can pass every formatting check and still not correspond to any real postcode — for instance, a plausible-looking combination that was simply never assigned, or one that’s been decommissioned. If accuracy matters (checkout addresses, delivery forms, compliance-sensitive data), format validation alone isn’t enough; it should be paired with a genuine postcode lookup.

Common UK Postcode Mistakes

A few errors come up repeatedly, whether from manual entry or careless form design:

Wrong space placement — inserting the space based on visual habit rather than the fixed three-character inward code.
Dropping a character — especially easy to do with longer outward codes like DN55 or EC1A.
Adding extra characters, often from copy-pasting an address that includes stray punctuation.
Confusing similar characters — particularly 0 (zero) and O, or 1 and I, which is part of why certain letters are avoided in the format altogether.
Assuming a single fixed pattern — building validation around only one of the six shapes (commonly AA9 9AA) and rejecting perfectly valid postcodes that follow a different pattern.
Treating a pattern match as proof of validity — as covered above, this is a format check, not confirmation the postcode exists.
Submitting a partial postcode — the outward code alone (e.g. SW1A) where a complete postcode is required.
Using an outdated postcode — postcodes do occasionally change or get retired, particularly in areas that have been redeveloped.
Assuming one postcode equals one property — it usually covers several.

UK Postcode Format and Online Forms

Postcodes show up constantly in checkout pages, delivery forms, registrations, government services, and mapping tools — and how strictly they’re validated varies by system. Some points worth knowing if you’re filling in (or building) these forms:

Many systems automatically normalise capitalisation and spacing, so sw1a1aa and SW1A 1AA are usually treated as equivalent.
Some forms perform only a format check; others perform a full lookup against a postcode database and will reject anything not actually on record.
Address-lookup tools that auto-populate a property list from a postcode are doing real validation, not just format checking — that’s why they can return “no results found” for a postcode that looks structurally fine.
Because validation rules aren’t standardised across every website, a postcode accepted on one form can occasionally be rejected on another, particularly for newer developments not yet reflected in an older postcode database.

UK Postcode Format vs Other Countries

The main structural difference between UK postcodes and most other postal-code systems is that the UK format is alphanumeric and hierarchical by design, rather than a fixed block of digits. A US ZIP code, for example, is always five digits (optionally extended to nine with ZIP+4), while a UK postcode varies from five to seven characters and mixes letters and numbers throughout, encoding area, district, sector, and unit in a single string. This is part of why UK postcode validation logic can’t simply reuse a numeric-only pattern built for other countries’ formats.

Troubleshooting: Why Is My Postcode Being Rejected?

If a form keeps rejecting a postcode you’re confident is correct, work through these in order:

Check the spacing — make sure there’s exactly one space, in the position that leaves the last three characters as the inward code.
Check for typos — particularly 0 vs O and 1 vs I.
Confirm it’s the complete postcode — not just the outward code.
Check you’re using the current postcode — especially for newly built addresses, which can take time to appear in some databases.
Consider that the postcode may be correctly formatted but not actually assigned — a format match isn’t the same as a real, active postcode.
Remember the website’s validation rules are its own — some are stricter (or older) than others, so a rejection on one form doesn’t necessarily mean the postcode itself is wrong.

Quick Recap

Every standard UK postcode splits into an outward code (area + district, 2–4 characters) and an inward code (sector + unit, always exactly 3 characters), separated by one space.
There are six main format patterns, all following the same letters-then-digits-then-space-then-digit-and-two-letters logic.
A postcode narrows down to a small group of addresses, not a single building — a full address still needs a building number, street, and post town.
Matching a valid format pattern is not the same as being a real, assigned postcode; genuine validation requires checking against an actual postcode database.