Hampshire's councils hold 37,474 business premises. Not one names a sole trader.

We loaded the published business rates list from every Hampshire council that publishes one — 37,474 rated premises. Of the 3,711 creative businesses we hold in those areas, 209 appear as a named ratepayer. Of the 91 we identify as sole traders, none do. Not because the councils don't know: because the law stops them saying.

12 August 2026 · by Rosie

Every council in England bills business rates, and most of them publish the list. Not a summary — the actual rating list, one row per rated premises, with the address, the rateable value, what the Valuation Office calls it and who is liable to pay. They publish it to stop having to answer the same Freedom of Information request every month.

This is, on paper, the most complete public record of commercial premises that exists. So it is worth asking a blunt question about a directory like this one: are we just rebuilding something the council already has?

We loaded every Hampshire rating list we could find published — nine councils, 37,474 rated premises — and matched it against what we hold. The answer turned out to be more interesting than yes or no.

What we found

Not one. Not a small number, not a matching problem we could tune away — zero out of ninety-one, across nine councils and thirty-seven thousand premises.

Why, and it isn't incompetence

The councils are not missing these people. They are legally barred from naming them. Southampton says so on the page next to the download link: the ratepayer's name and correspondence address are withheld where the ratepayer is an individual — sole traders, partnerships and so on — because that is personal data, exempt under section 40(2) of the Freedom of Information Act 2000. Winchester uses almost the same words. So does East Hampshire.

Which means the blank rows in these files are not gaps. They are the sole traders, present as an outline: an address, a rateable value, a description of the premises, and a legally-required silence where the name goes.

Three councils go further and delete the row

Basingstoke and Deane, Rushmoor and Test Valley publish files with zero withheld rows, and it would be easy to read that as "no sole traders here". It isn't. They have taken the individuals out of the file altogether — Test Valley says outright that it publishes only accounts where the ratepayer is a company.

The distinction matters more than it sounds. Where a council blanks the name you can still count the premises, see the rateable value, and know that somebody unincorporated is working there. Where the row is removed, that business has left no trace at all — you cannot even count it. Between them those three councils account for 9,201 premises in which an unknown number of sole traders are invisible twice over.

Four different conventions for the same legal obligation, across nine neighbouring councils. Anyone trying to compare one district with the next has to notice that first, or they will conclude that Basingstoke has no self-employed people in it.

The withheld businesses are the small ones

The rateable values make the pattern plain. Of the withheld premises carrying a value, 79% fall under £12,000 — the threshold below which small business rate relief wipes out the bill entirely. Among the named ratepayers, it is 44%.

So the population being withheld is not a random slice. It is systematically the smallest premises, run by people who pay no business rates at all, in exactly the size band that local growth funding is usually written for.

What this means if you allocate money

Funding for creative and small business support is allocated by council area, and the allocation has to be justified with a number. The rating list is a natural place to look for that number. It will give you a defensible count of companies with premises, and it will silently omit the self-employed.

• A count of creative businesses drawn from the rating list is a count of incorporated creative businesses with premises. It is not a count of the sector.
• The gap is not evenly spread: it is largest exactly where one-person businesses cluster, and it disappears entirely in the three districts that remove the rows.
• Nobody is doing anything wrong. Section 40(2) is the right law and the councils are applying it correctly. The measurement gap is a side effect, and it needs filling from somewhere else.

That is what this directory is for. We find these businesses from the sources that *are* allowed to name them — the ICO's register of fee payers, business listings people publish themselves, their own websites — and attribute each one to a council area. It is the same population the rating list cannot show you, arrived at from the other direction.

What the rating list is genuinely good for

Having said all that, it is not a dead end. Three things it does well, which we now use it for:

• Confirming premises. A company's registered office is often its accountant's address. A rated premises is somewhere work actually happens.
• Finding the creative venues. 116 premises across Hampshire carry a Valuation Office description that is genuinely creative — recording studios, galleries, theatres, dance schools, a handful of museums. That is a small, checkable list.
• Mapping where makers sub-let. Where one landlord is the ratepayer for forty units, the people inside are sub-tenants and appear nowhere. Those buildings are the highest-density places to go and find businesses in person — which is a far better use of an afternoon than another search query.

How this was done

Nine council releases, downloaded from the pages linked below, loaded into one table and attributed to a council area by postcode using the ONS Postcode Directory. No two councils use the same column names, so the loader maps each authority's spellings onto a common set of fields.

Matching a rated premises to a business we hold requires the normalised name and the postcode to agree. Name alone is not enough at this scale — there are several firms called Southampton Something Limited — and postcode alone is worse, because an office block shares one postcode across forty tenants. That strictness is why 5.6% is a floor rather than an estimate: some real matches will be missed by a spelling difference. It cannot explain a zero out of ninety-one.

Figures are as at the release dates in the table above, which range from April to August 2026 depending on how often each council publishes. Most publish monthly; Fareham and Havant quarterly.