A guide to using ONS and Companies House data for UK stakeholder maps

What to take away

• Corporate communications teams should build UK stakeholder maps from named ONS datasets and Companies House records, not from generic personas.
• Pull ONS business, industry and trade statistics for sector context, and ONS population data for audience composition.
• Use Companies House confirmation statements, accounts filings, and People with significant control data to identify who controls and influences a company.
• Cite ONS methodology when quoting statistics so the map survives scrutiny from regulators and internal reviewers.
• Keep register facts separate from claims in the map, and label each entry as sourced or inferred.

Why stakeholder maps need records, not personas

A persona says "finance director, 45, cares about efficiency". A record says "company X filed accounts showing turnover of £12m, and its confirmation statement names two PSCs". The second is defensible in a board meeting. The first is not.

UK communications strategy leads are increasingly asked to show evidence. Regulators including the FCA, the Takeover Panel, and the Charity Commission for England and Wales expect organisations to know who their stakeholders are. A map built on register data answers that expectation.

Generic personas also hide regional variation. A map for Greater London looks different from one for the North West, Scotland, or Wales. Records reveal those differences because they carry addresses, sectors, and filing histories.

Start with populations, not channels, when you scope the map. That approach is set out in our guide to audience research for corporate communications that starts with populations, not channels. It keeps the work focused on who is affected before you decide how to reach them.

The named ONS datasets worth pulling first

The Office for National Statistics publishes two families of data that matter for stakeholder mapping. The first is business, industry and trade statistics, covering UK business counts, turnover, and sector structure. Use it to size a sector and find where companies cluster. The ONS business, industry and trade statistics are the starting point.

The second is people, population and community data, covering demographics, migration, and cultural identity. Use it to describe the audience around a site, a region, or a customer base. The ONS people, population and community data hold the relevant releases.

Within those families, name the datasets you actually cite. UK Business Counts gives counts of enterprises by size and region. The Annual Business Survey provides turnover and approximate gross value added by sector. Business Demography covers births, deaths, and survival of businesses.

Population estimates by local authority underpin audience sizing. Census 2021 data on national identity, language, and religion supports cultural identity work in England and Wales. Labour Market Statistics give employment and occupation patterns.

ONS methodology pages explain how each dataset is constructed, including sampling, revisions, and definitions. Cite that methodology alongside any figure you lift. It is the difference between a credible map and a decorative one. The ONS methodology pages are the reference to link.

Pull data at the geography that matches your decision. Local authority data suits regional campaigns. Combined authority and region data suit devolved or mayoral contexts. UK-level data suits national strategy. Do not mix geographies in one table without labelling them.

Companies House filing data: confirmation statements and accounts

Companies House holds the UK company register. It is free to search at a basic level, and filings are public. The GOV.UK service for getting information about a company explains what is available and how to request it.

Two filing types do most of the work for stakeholder mapping. The confirmation statement confirms the company's registered details, including its registered office address, directors, and shareholders. It is filed at least annually.

The accounts filing gives financial statements, and for small companies often just a balance sheet. Together they show who the company is and how big it is.

Use confirmation statements to track changes over time. A change of registered office from one region to another is a signal worth noting. A new director with a background in a regulated sector is another. These are the UK records that should prompt corporate communications research.

Accounts filings help you segment stakeholders by size. Turnover bands, employee numbers where disclosed, and accounting basis (micro-entity, small, full) all indicate capacity and likely interest. A micro-entity rarely has a communications function. A full-list company almost certainly does.

Filing history also shows compliance behaviour. Late filings, amended accounts, and gazette notices are public. They do not prove anything about reputation, but they are facts you can log.

People with significant control as ownership evidence

People with significant control (PSC) data is the closest the UK register gets to an ownership map. A PSC is a person who meets one or more conditions, such as holding more than 25% of shares or voting rights, or having the right to appoint or remove a majority of the board.

The GOV.UK guidance on people with significant control explains the conditions and the register.

For stakeholder mapping, PSC data answers a question personas cannot: who ultimately decides. If a company is owned by a holding company registered overseas, the PSC register may name a person or may record a relevant legal entity. Either way, it tells you where influence sits.

Use PSC data to build ownership chains. A UK subsidiary may have a PSC that is itself a company. Follow the chain until you reach a person or a state-owned entity. That chain is often the real stakeholder map for a regulated decision.

PSC data also supports conflict and interest checks. If two companies in your map share a PSC, they are connected. That connection may matter for consultation responses, media briefings, or coalition building.

Treat PSC data as evidence, not as a complete picture. Some entities are exempt, and some filings are incomplete. Note the gaps rather than filling them with assumption.

Building a UK audience map from ONS population and cultural identity data

An audience map starts with who lives and works in a place. ONS population estimates by local authority give the base. Age structure, household composition, and migration patterns shape how a message lands.

Cultural identity data adds nuance that population counts miss. Census 2021 national identity, language, and religion data for England and Wales show how people describe themselves. That matters for messaging in the West Midlands, Greater London, or Wales, where identity patterns differ sharply.

Devolution means you cannot treat the UK as one audience. Scotland, Wales, and Northern Ireland have their own health, education, and some regulatory contexts. ONS publishes some data on a UK basis and some on England and Wales. Check the coverage before you cite.

Labour market data links audience to sector. If your stakeholder map is about a workforce issue, employment by occupation and industry tells you which local authorities have the most at stake. That is more useful than a demographic persona.

The aim is to map who a corporate communication affects, not just who receives it. Population and identity data feed that map at the top, before any channel decision is made.

Worked example: a regional stakeholder map for one sector

Take a hypothetical UK communications team at a mid-sized energy supplier. The brief is to map stakeholders for a proposed grid connection in the North West of England. The team has two weeks and no budget for a survey.

Step 1. Define the decision and the geography. The decision is a planning application. The geography is the local authority area plus the combined authority and the North West region.

Step 2. Pull ONS business statistics for the sector. Use UK Business Counts to find how many energy-related enterprises operate in the region. Use the Annual Business Survey for turnover context. Cite ONS methodology for each figure.

Step 3. Pull ONS population and identity data for the local authority. Population estimates give the number of residents. Census 2021 data gives language and national identity profiles that shape consultation materials.

Step 4. Build the company list from Companies House. Search for companies with registered offices in the area that are in the supply chain. For each, download the confirmation statement and latest accounts. Log registered office, directors, size band, and filing dates.

Step 5. Add PSC data. For each company, record the PSC and any ownership chain. Flag companies that share a PSC, because they may act together in consultation.

Step 6. Classify each entry. Use four labels: decision maker, influencer, affected party, and observer. A local authority planning committee is a decision maker. A neighbouring landowner is an affected party. A trade association is an influencer. A national media outlet is an observer.

Step 7. Test the map. Ask whether each entry has a source. If not, mark it as an assumption and set a task to verify it. The map is a working document, not a finished report.

The output is a table with columns for name, type, source, and evidence. Here is a simplified version.

Stakeholder	Type	Source	Evidence
Local authority planning committee	Decision maker	ONS population data	Resident population and ward boundaries
Regional energy company A	Influencer	Companies House	Confirmation statement and accounts
Landowner B	Affected party	Companies House	PSC register entry
Trade association C	Influencer	ONS business statistics	Sector enterprise counts
National media outlet D	Observer	Public record	Ofcom licence records

That table is the map. Every row can be defended because every row has a source.

Keeping register facts apart from claims in the map

A register fact is something Companies House or the ONS states. A claim is something you infer. Mixing them weakens the map.

Use a simple convention. Label each entry as either "sourced" or "inferred". Sourced entries carry a link to the filing or dataset. Inferred entries carry a note on the reasoning and a date.

Review the map before it leaves the team. Check that no inferred claim is presented as fact. Check that every statistic has a methodology link. Check that no figure is older than the decision it supports.

This discipline is the same one we recommend in our guide to competitor research for corporate communications: register facts apart from claims. It applies to stakeholder maps because both rely on public records.

Finally, set a review date. Company registers change. ONS releases are revised. A map built in September may need refreshing after the next accounts season. Put the date in the document.

Common questions

How often should I refresh a stakeholder map built from Companies House data? At least annually, after the main accounts filing season. Refresh sooner if a company in the map files a change of directors or registered office.

Can I use ONS data for a Scotland-only stakeholder map? Yes, but check coverage. Some ONS releases cover England and Wales only. For devolved matters, use the relevant Scottish Government statistics alongside ONS.

What if a company has no PSC listed? Record that as a gap. It may be exempt, or the filing may be incomplete. Do not infer ownership without evidence.

Do I need a licence to reuse Companies House data? Companies House data is generally available for reuse under an open licence. Check the current terms on GOV.UK before publishing extracts.

How do I cite an ONS figure in an internal map? Link to the dataset release and the ONS methodology page. State the geography and the period the figure covers.

Is a persona ever useful? As a communication device, yes. As evidence in a stakeholder map, no. Keep personas separate from the sourced map.