Why filing-traced data beats scraped profiles for UK sales

Data traced to a statutory filing has a named source, a date, and a legal duty of accuracy behind it. Scraped profile data has none of those. For UK sales, that difference decides bounce rates, compliance and trust.

Data traced to a statutory filing has three things scraped profile data never has: a named source, a date, and a legal duty of accuracy standing behind it. When a UK company files with Companies House, a real person is confirming facts on a public document, and filing false information is an offence. When a profile is scraped, nobody confirmed anything, on any date, with anything at stake. For UK sales, that difference is not philosophical. It decides your bounce rate, your compliance posture and whether a prospect's first impression of you is "accurate" or "where did you get this?"

What filing-traced means

Filing-traced data is data where every field can answer the question "where did this come from, and when?" with a specific document. A company's name, status, incorporation date, registered office, SIC codes and officers all originate in filings on the public record, each one dated and attributable. Leadistry is built on exactly this standard: every record starts from the company's Companies House record, and what we add through enrichment, the website, the business email, is discovered from the company's own published sources and verified before delivery. Our data sources page and methodology document the chain field by field.

Scraped profile data inverts the model. It starts from what people write about themselves on social platforms and directory sites, harvested in bulk, with no visibility of when a fact was true or who asserted it.

Provenance, freshness, accountability

The two models differ on three properties, and each one shows up in day-to-day selling:

• Provenance. A filing names its company and its date. A scraped field is an orphan: you cannot tell whether "Head of Operations at Acme" was scraped last week or three years ago, or whether it was ever true. When a prospect asks where you got their details, "from your company's public filings and your own website" is an answer you can give out loud.
• Freshness. Filings are events with dates, so filing-based data can be refreshed on a schedule and shown to be current. Profiles decay silently: people change jobs and never update pages, companies dissolve and their profiles trade on. Nothing marks a scraped record as stale, which is precisely what makes it dangerous.
• Accountability. A confirmation statement is a director telling the state, on the public document, that the company's details are correct, with consequences for getting it wrong. No scraped bio carries a duty like that. One dataset has an accuracy incentive built into its source; the other has an engagement incentive.

How scraped data fails in practice

The failure is rarely dramatic. It is a slow tax:

• Bounces. Emails guessed from stale profiles bounce, and mailbox providers read bounce rates as a spam signal, so bad data quietly burns the sender reputation your outreach depends on.
• Misdirected personalisation. Opening lines built on a three-year-old job title impress nobody. The prospect knows their own history; a wrong fact reads worse than no fact.
• Ghost companies. Scraped databases keep selling contacts at companies that dissolved, merged or renamed. The public record settles a company's status definitively; a profile does not.
• Unanswerable questions. Under UK GDPR, individuals can ask what you hold about them and where it came from. "A vendor scraped it from somewhere" is an uncomfortable answer for you, and a worse one for your vendor. We take the compliance side further in our UK B2B data compliance guide.

We compared the two source models head-to-head for prospecting in Companies House vs LinkedIn.

Where profile data still helps

Honesty requires the other column. Profile and social data shows things filings never will: what a company says it is up to right now, who is publicly active in what role, the tone of the shop window. Used as colour on top of a verified record, it sharpens a message. The failure mode is using it as the record: building the list, the identity and the contact details from scraped profiles alone, with nothing underneath. Filing-traced data is the skeleton; let profiles be, at most, the paint.

What tracing looks like in Leadistry

Concretely, for every delivered lead:

• The company's identity, status, age, SIC codes and officers come from its Companies House record, refreshed daily.
• The website is discovered and validated as belonging to that specific company, not a namesake, and the no-website result is itself a verified data point.
• The business email comes from the company's own published sources and is verified before it counts as delivered; anything that fails verification never reaches you or your allowance.
• Every delivered lead is re-verified monthly, so a company that dissolves or a site that vanishes does not sit in your pipeline as a ghost.
• Every company has a permanent removal route, and suppressions hold account-wide.

That chain is also why we publish comparisons against tools built the other way: the difference is checkable, field by field.

The test to apply to any data source

Ask two questions of any UK lead list. First: for each field, can the vendor tell you the document or page it came from and the date it was true? Second: when a fact changes in the real world, what causes the record to change, a scheduled re-check against a dated source, or nothing until a customer complains? A vendor with good answers to both can be audited, defended and built on. A vendor without them is renting you their guesses, and the rent comes due in bounces and awkward replies. Start free with 10 leads and trace every field yourself.