One London media buy covers Richmond, on median pay of £1,056.60 a week, and Barking, on £754.10.
Geo targeting usually means a radius around a store, or a region someone picked from memory. The ONS publishes the actual geography of the country in units built for this job, with income, occupation and business density attached. The series hub maps the full set.
The units
England and Wales divide into 188,880 areas of a few streets each, nesting up through small neighbourhoods and districts to the boroughs you already know.
The smallest published area covers a few streets: 100 to 625 people, and England and Wales divide into 188,880 of them. Above it sits a small neighbourhood of 1,000 to 3,000 people, then a district of 5,000 to 15,000, then the local authorities you already know.
Every address in England and Wales sits inside a published ladder that ends at a few streets.
Then there is the commuting area. There are 228 of them in the UK, drawn so that at least 75% of the people who live in one also work in it, relaxed to two thirds for the largest urban areas. That is a catchment defined by observed behaviour. The commuting map is a better first draft of a store’s catchment than a circle with a five-mile radius drawn from the door.
The layers
Income, rurality and deprivation are published for every small area, and the free postcode directory joins a sales file to all of it in one join.
Census Maps puts any census variable on a zoomable map, down to a few streets, in a browser. The small-area income estimates model household disposable income for every district in England and Wales, and the range runs from £20,800 to £107,600. Of the 50 highest-income areas, 47 are in London.
The highest-income neighbourhoods are almost all in London.
Of the 50 highest-income small areas in England and Wales, 47 sit inside the capital.
Estimated. Note 3. Small-area income is modelled with wide intervals, so read the concentration, not the ranking. · gabrielgruter.com
The rural-urban classification flags every area as urban or rural instead of leaving it to placename guesswork, and the English Indices of Deprivation got their first refresh since 2019 in October 2025. The join key for all of it is the ONS Postcode Directory, a free quarterly lookup from every UK postcode to every unit above. Feed it a postcode-level sales file and your first-party data becomes census geography in one join.
A worked contrast
Richmond upon Thames has 78.2% of its working residents in the top three occupation groups, on median weekly pay of £1,056.60. Barking and Dagenham, in the same city and often in the same media buy, has 39.8% on £754.10.
The same media buy reaches two different labour markets.
A Richmond upon Thames full-time worker
£1,056.60median weekly pay
- Working residents in the top three occupation groups78.2%
A Barking and Dagenham full-time worker
£754.10median weekly pay
- Working residents in the top three occupation groups39.8%
A premium brand running one London budget with one message is treating those two places as one.
The same logic runs for B2B. The business register serves enterprise counts by size, industry and local authority, down to five-digit industry codes, and the B2B sizing piece puts that table to work.
Landing it in platforms
No ad platform can target a district directly, so the data lands as separate campaigns by area, budget splits by borough, and store catchments drawn by commuting area.
Districts are not a targeting unit on any ad platform, so the layer lands at campaign level. Split campaigns by area and give each area its own target. Run borough budget splits as geo experiments. Where the platform allows it, use conversion-value rules by location so the bidding learns the geography. And define store catchments by commuting area before anyone draws a radius. The suits brief runs this playbook on one category.
Where this argument is weakest
The income figures are modelled, the travel data is pandemic-era, and a rich area is not a rich person. Weight with the layers; never exclude on them alone.
The census travel-to-work data was collected mid-pandemic, and the ONS’s own guidance says to take care not to overinterpret it, which is why the 2011-based commuting areas remain the recommended set. The small-area income figures are model-based estimates, not measurements. The deprivation indices cover England only and belong to MHCLG rather than the ONS. And the oldest trap in geography is still the ecological fallacy: a rich area is not a rich person, and a postcode tells you very little about the individual behind it. Weight with these layers. Do not exclude on them alone.