Postal geographies

Royal Mail maintains a UK-wide system of postcodes to identify postal delivery areas.

Most people know their postcode, so we are able to use this as the main geographic reference when collecting data.

This reference can be related to any geographic unit used for statistical production, such as a local authority district or electoral ward.

So postal geography is very valuable.

Postcode structure

Postcodes are alphanumeric references comprising an outward code of 2–4 characters and an inward code of three characters. For example:

The postcode is structured hierarchically, supporting four levels of geographic unit:

As at August 2022, these 1.79 million live postcodes comprise approximately 1.7 million small user and 0.09 million large user postcodes, including the Channel Islands and the Isle of Man.

Unit postcodes

Unit postcodes are the base unit of postal geography and fall into two types:

• Large user postcodes: allocated to single addresses receiving at least 500 mail items per day (e.g. business addresses)

• Small user postcodes: collections of (usually) adjacent addresses. A single small user postcode may contain up to 100 addresses, but 15 is a more typical number

Note: it is possible for large buildings with many separate delivery points (for example, a tower block) to have more than one unit postcode within the building.

Limitations of using postcodes as a geographic reference

Postcodes form a compact geographic reference that the public and businesses are familiar with.

Linking postal geographies to other geographic units is far from straightforward though, as postcode boundaries mostly do not align with other geographic boundaries. If a postcode boundary (reflecting all the addresses with that postcode) straddles a ward (or other) boundary, you have to decide which ward to allocate any data for that postcode to. Our postcode directories take the grid reference of the postcode centroid and match this up to digital administrative boundaries. However, some addresses (and therefore data) will still inevitably be allocated to the "other" area. Note though that this problem will be reduced in future with the move towards using address-based rather than postcode-based grid references.

Also, postcode boundaries are subject to continuous change due to new addresses, single addresses acquiring large user postcodes as mail volume increases, and the need to restrict the number of addresses per unit to less than 100. Areas can also be recoded and postcodes can be re-used in a different place after just two years. Continuous monitoring is therefore required to avoid data misallocation.