Companies House Filing Monitor - UK Company Changes

SOURCE COVERAGE & PRACTICAL LIMITS

Full product guide

Documentation captured from build 1.0.4. Example outputs retain their original sample context; they are not live results.

Company filings, ready for review

Maintain a sourced filing feed for supplier, customer or company research. This Actor reads bounded public Companies House filing-history pages and can compare saved filing metadata over time.

Open this Actor in Console · First run · Results · Cost · Full guide

Try this Actor

1. Choose a small scope. Use the existing Tesco filing feed example below: company 00445790, one filing-history page and at most five saved rows. Keep comparison off for a first look.
2. Run and review. Open the dataset for filing observations and the run's key-value-store SUMMARY for each company's page-window coverage. A company page failure does not produce paid partial-window rows.
3. Use the results. Export CSV for a filing list or JSON for comparisons. Follow the official document link when you need the underlying filing; this Actor does not download PDFs.

Read this before using the data: The filed date is not necessarily the transaction or effective date mentioned in a document. A bounded filing list is not a full archive or a legal/compliance verdict. The Actor does not send alerts itself.

Tesco Company Filing History — copy this input into the JSON editor in Console.

{"companyNumbers":["00445790"],"maxResults":5,"maxSourcePages":1}

Result preview

The dataset view highlights these columns:

• Filing ID · filingId
• Company number · companyNumber
• Current company name · companyName
• Filed date · filedDate
• Filing type code · filingType
• Published description · description

Use the original source links and the field definitions below when interpreting the results. CSV works well for flat columns; JSON preserves nested values.

Usage and pricing

Base result rate on the FREE tier: $3 per 1,000 company filing observations. The Pricing tab shows your plan's result rate and any Actor Start charge. Set a small result limit for the first run; the requested limit is not a guarantee of that many results.

Full guide

Monitor a small, explicit UK company watchlist for new filings and changed filing descriptions or annotations. Export stable filing IDs, filed dates, type codes and official document links. The Actor uses public Companies House pages and needs no API key.

Why use it

Connect a supplier or customer watchlist to a repeatable filing feed. Each saved observation includes a stable transaction ID and a content hash. Optional named history identifies new or changed filings and provides the previous observation and changed fields. Use Apify integrations to consume the dataset; this Actor does not send alerts itself.

Price: $0.003 per saved filing on FREE/BRONZE, plus the configured $0.00005 start event. SILVER results are 20% less; GOLD and higher are 40% less. Unchanged observations skipped by onlyChanges have no result charge. Apify's configured prices govern the actual charge.

Fast default

{} reads one Tesco filing-history page and saves at most five filings. companyNumbers is an explicit watchlist, not a search or bulk export. A company number can be eight digits or a two-letter prefix with six digits; numeric strings are padded with leading zeros.

Inputs

The date/type filters inspect only the selected pages. Older matching records can be outside that window. The whole selected company window is fetched and validated before that company's results are emitted. A page failure makes that company incomplete and produces no paid partial company-window observations.

Output

One row per saved filing:

• filingId (companyNumber:transactionId), transactionId, companyNumber, companyName.
• filedDate, filingType, and description as published, including annotation text visible in the index.
• documentPages where the source publishes a page count; otherwise null.
• url (official document link), sourceUrl, sourcePage, dateMeaning, contentHash, observedAt and attribution.
• comparisonStatus: not_requested, new, changed or unchanged; changedFields, previousObservedAt, and previous.

Date meaning matters: filedDate is when the document was filed. A description may contain earlier transaction or effective dates. Those are preserved as text and never silently substituted for the filing date.

The content comparison covers filed date, filing type, description/annotations and page count. It does not compare PDF contents. A later company-name change updates the displayed current company name but does not label every historical filing as changed.

Three exact examples

Tesco filing feed

Use the copyable first-run input above.

Unilever accounts and confirmation statements

{"companyNumbers":["00041424"],"filingTypes":["AA","CS01"],"maxResults":5,"maxSourcePages":1}

BP filing feed

{"companyNumbers":["00102498"],"maxResults":5,"maxSourcePages":1}

Repeated observations

History is kept in the named key-value store maydit-companies-house-filings-history; snapshotName, company list, source-page window and filters define its scope. The first comparison run emits new observations and logs a warning. onlyChanges can legitimately return zero rows when the selected observations are unchanged.

History advances only for confirmed saved rows. A rejected spending-limit row never updates history. A final accepted row still counts if the SDK also signals the spending cap. If the saved result limit is smaller than the selected source window, a later onlyChanges run can emit previously unsaved filings from that same window. To establish a complete one-page comparison baseline, use one company and maxResults:1000 for both baseline and repeat runs.

Before committing, the Actor re-reads history and rejects a detected concurrent change. This is an optimistic check, not an atomic lock: run one writer per snapshot scope. Malformed history is rejected, and a new snapshotName provides a clean baseline. Missing filings never generate deletion events; a bounded index does not prove a filing was removed.

Coverage and failures

SUMMARY reports status, incomplete, rowsSaved, rowsObserved, unchangedSkipped, sourcesSucceeded, sourceCoverage, failedSources, and limit flags. The normal bounded page or result cap is intentional scope, not a full company archive. An upstream, parser or deadline failure is explicit; if no usable observations can be saved, the run fails. Valid empty filters, unchanged repeats and a zero-result spending stop are reported without fake result rows.

The Actor does not search the company database, download PDFs, extract accounts, retrieve officer profiles or private contact details, or determine legal/compliance status. Public filing descriptions can name officers in their professional role. Company and filing facts are recorded as published, not independently verified.

Source and access

Companies House guidance states that the public service is available without registration. Its searchable databases are not intended for bulk downloads and may restrict high access volumes. This Actor uses bounded watchlists, checks robots rules, paces requests and stops on denied access. It does not bypass an API key: the separate official API requires credentials, while this product reads public HTML pages.

Source markup, temporary outages and shared cloud rate limits can change. A successful sample is not an availability guarantee. Contains public sector information licensed under the Open Government Licence v3.0.

For support or correction/deletion requests, use Actor Issues or bmay@mayd-it.com.

Help with your run

Open the Issues tab on this Actor with the run URL, the input field that caused trouble and what you expected to receive. Remove API keys and private information before posting.

Explore more Mayd-It Actors

Understand your run

Open Run report in the run’s Output tab for saved-record counts, reported source issues and limits, and suggested next steps. Use CSV export to take the saved results into a spreadsheet. The existing JSON results and detailed machine-readable summary remain available.

The report reflects this run’s recorded coverage. A successful run or a saved result does not establish that every possible source record was collected.