How does UK postcode structure work?
Royal Mail's postcode system is a hierarchy, not a flat lookup. Understanding each level is the starting point for any geography-first consumer campaign, because the level you choose determines how precisely you can carve out a catchment area and how many records you can realistically work with.
The four levels are nested inside each other, with each level adding one more component to the postcode string:
| Level | Example | Structure | Approx. count in UK | Avg. households | Best use case |
|---|---|---|---|---|---|
| Postcode area | SW | 1 or 2 letters | 124 areas | 1.5M+ | Region-wide brand campaigns, broad TV/OOH geographic planning |
| Postcode district | SW1 | Area + 1 or 2 digits (sometimes a letter) | ~3,000 districts | ~60,000 | City-quarter campaigns, large retail catchment areas |
| Postcode sector | SW1A 1 | District + space + 1 digit | ~11,000 sectors | 2,500–4,000 | Store radius targeting, direct mail catchment work, test-cell design |
| Unit postcode | SW1A 1AA | Sector + 2 alphanumeric chars | ~1.8M active | ~15 | Micro-targeting, follow-on campaigns, geolocated door drops |
These are approximations. Urban postcodes are denser; a unit postcode in central Manchester may cover 80 or 90 flats. Rural unit postcodes sometimes identify a single farm. The average of 15 deliverable addresses holds nationally but can be misleading in either direction once you leave suburban England.
Which postcode level should you select for consumer campaigns?
Postcode area: regional brand campaigns only
With 124 areas covering all of Great Britain, the area level is too coarse for most response-driven campaigns. A single area like SW covers vast swathes of inner and outer London. It is useful when you are planning regional TV or billboard activity and want your direct mail to reinforce the same geography, or when you genuinely want blanket penetration across a major city. For anything more surgical, go a level deeper.
Postcode district: the city-quarter sweet spot
Districts work well for campaigns with a city-quarter logic: a legal firm in Leeds City Centre (LS1) wanting to reach nearby businesses and households, or a restaurant group covering Brighton's BN1 catchment. Each district holds around 60,000 households on average, so a single district selection gives you volume worth mailing without producing an unwieldy list. The boundary precision is still limited, though. LS1 and LS2 do not neatly map to the kind of drive-time radius that retailers use for catchment planning.
Postcode sector: the campaign workhorse
Postcode sector is where most consumer campaign planners spend most of their time, and for good reason. At 2,500 to 4,000 households per sector, a selection of 20 sectors gives you a list of 50,000 to 80,000 records: enough to run a statistically valid direct mail test with decent split-cell sizes. That same granularity fits neatly onto a drive-time catchment map around a retail location. In our experience, sector-level selections are the most productive starting point for new postcode-targeted campaigns because they balance coverage, precision, and the minimum volumes needed to draw meaningful conclusions from response data.
Sector selections also compose cleanly. If you are building a store-radius selection around a new branch, you can list the sectors whose centroid falls within a 10-minute drive time, rather than drawing an imprecise circle across district boundaries.
Unit postcode: precision follow-up
With around 15 addresses per unit postcode, targeting at unit level is not really about geography any more. It is micro-targeting: a house-by-house selection driven by geodemographic classifications, individual-level data, or a hand-curated set of addresses. This is the right level for precision door-drop campaigns using Royal Mail's Door to Door service where you are specifying individual postcode walks, or for follow-on campaigns to respondents from a previous mailing where you want to suppress entire postcodes where response was strongest (to avoid over-saturation) or mail only postcodes where you know you have conversion data.
How do geodemographic classifications fit in?
Geography tells you where someone lives. Geodemographics tells you what kind of household they probably are. The two work together.
Geodemographic classification systems assign every unit postcode in the UK to a lifestyle and socioeconomic segment, derived from a combination of census data, credit and financial behaviour, retail spending patterns, property values, and other aggregated signals. They produce category names (the specific names vary by supplier) that describe household types: affluent urban professionals, retired rural owner-occupiers, young renters in inner-city flats, families in new-build estates, and so on.
These classifications are attached at unit-postcode level. That means you can express a selection like "all households in the SW postcode area that fall into the top two socioeconomic tiers, excluding flat-dwellers". Without the geodemographic overlay, you would have to mail every postcode in SW and accept the wastage. With it, you can drop the sectors dominated by high-density social housing and focus budget on the sectors that match your customer profile.
Geodemographic codes and individual-level data work differently
A geodemographic code is a property of the postcode, not the individual. Two households in the same unit postcode share the same geodemographic classification regardless of their personal financial situation. This is worth understanding because it affects how you layer your selection criteria.
When you buy from a fully opt-in consumer file, you can combine postcode-level geodemographic codes with individual-level data such as declared age, estimated household income band, property ownership status, number of children at home, and stated interests from questionnaire responses. That combination is significantly more precise than either dimension alone. A postcode-only selection finds the right street; adding individual-level age and homeowner status finds the right household on that street.
Combining postcode selections with demographic criteria
The practical workflow for a postcode-led consumer campaign looks like this. You start with a geographic brief (radius around a store, a set of postcodes from a catchment analysis, a county or region). You apply that as a filter to the consumer file. Then you layer on demographic criteria to reduce the count to the households most likely to respond.
Typical stacked selections include:
- Age band: 35 to 64 is the core direct mail audience for most financial, home improvement, and insurance products; 55-plus for retirement-related categories.
- Property ownership: owner-occupier flag is the single most reliable predictor for home improvement, equity release, and utility-switching campaigns.
- Household income band: available as a modelled estimate on most consumer files, not a declared figure. Treat it as a decile ranking, not a salary.
- Presence of children: useful for education, family holidays, childcare, and retail categories.
- Declared interests: hobbies and lifestyle flags gathered at the point of consent, such as gardening, DIY, personal finance, travel, or pets.
- Channel preference: some records carry channel opt-in flags (postal, email, telephone). Filtering to postal-opted records before a direct mail run reduces wastage and keeps you clean against channel preference records.
The order matters. Lead with geography to establish your universe, then apply demographics. If you reverse the order (start with all homeowners aged 55-plus in the UK, then filter to your postcodes), you get the same result mathematically but the count query is slower and the logic harder to explain when someone asks how the list was built.
Pitfalls in postcode-based consumer targeting
Boundary effects
Postcode boundaries do not follow natural catchment shapes. A 10-minute drive radius from a retail branch in Coventry will cut across postcode sector boundaries at an awkward angle. The sector whose centroid falls inside your radius may include thousands of addresses outside your true catchment, and a neighbouring sector whose centroid falls just outside may contain a dense pocket of households that would realistically shop with you.
The practical fix is to use a drive-time or walk-time polygon generated from a mapping tool, extract all unit postcodes that fall entirely within that polygon, and provide that list to your data supplier as an explicit inclusion list. This gives you precision at unit-postcode level without the wastage of mailing an entire sector to cover a partial overlap. It adds a step to the data brief, but the reduction in wastage typically justifies it for campaigns where the catchment boundary is genuinely important (food retail, leisure, local services).
Sparsely populated postcodes
Rural unit postcodes can mean a handful of addresses. If your sector selection includes large rural sectors in areas like the Scottish Highlands or mid-Wales, you may find that the sector contributes very few actual records because the physical sector covers a vast geographic area with low population density. The per-record cost stays constant, but the volume per sector is far lower than the national average suggests. Always pull a count by sector before committing budget, and flag any sector with fewer than 200 records as potentially uneconomic for analysis purposes.
High internal variation within a postcode
A geodemographic classification assigned to a unit postcode is a majority classification. It describes the dominant household type. Some postcodes have high internal socioeconomic variation: a row of Victorian terraces split between owner-occupiers and privately rented flats, or a postcode boundary that straddles a social housing estate and a private development built on the same former site. When the geodemographic code says "affluent professional household", it may only apply to 60% of the addresses in that unit postcode.
This is where individual-level data earns its keep. Rather than treating the geodemographic code as definitive, use it to identify candidate postcodes and then filter within those postcodes using individual-level attributes from the consumer file. That way you are mailing the homeowners in the postcode, not every address that happens to share the same classification code.
The PECR consent and MPS wash requirement
Postcode targeting is primarily a direct mail discipline, but the same geographic selections are often used for telephone and email follow-up. Before any channel goes live, check that each record carries the appropriate channel opt-in. For postal, wash against the Mailing Preference Service (MPS) before despatch. For telephone, wash against the Telephone Preference Service (TPS) every 28 days. The Privacy and Electronic Communications Regulations (PECR) require prior consent for B2C email and SMS regardless of how the geographic selection was built, so geographic targeting does not change the channel compliance obligations. See our guide on PECR rules for UK marketers for the full channel-by-channel breakdown.
What does SortedIQ's consumer file offer for postcode targeting?
SortedIQ's consumer file is a fully opt-in consumer file under UK GDPR and PECR consent, covering 10M+ UK records sourced from consumer surveys, lifestyle questionnaires, and similar consented channels. Every record carries an explicit opt-in to third-party marketing from the point of data collection.
Postcode selections are available at all four levels: area, district, sector, and unit. The most common request is a sector-based selection combined with one or two demographic overlays. We can also accept a client-supplied list of unit postcodes derived from a drive-time catchment analysis and match records precisely to those postcodes, rather than using the broader sector boundary as a proxy.
Geographic counts are available before purchase: tell us the postcodes you want, add any demographic criteria, and we will return a record count by postcode sector so you can see where the volume sits before committing. This is particularly useful for validating catchment assumptions before a campaign goes to print.
For more on what the full consumer file covers, see our UK consumer data overview (forthcoming) and consumer data pricing guide (forthcoming).