TL;DR
You need the name of whoever runs a company, not the switchboard. Directories won't give you that — they give you info@ and a category listing. The company's own website usually will, and it's usually on one of about eight pages.
Where owner and manager details actually live, in rough order of yield:
- Team / About / Leadership pages — the obvious one, and the highest yield
- The footer — registered company details, directors, registration numbers
- Legal pages — imprint, terms, privacy policy (a privacy policy names a data controller)
- Press / news sections — quotes are attributed with a name and title
- Careers pages — "report to" lines and hiring manager names
- Case studies and testimonials — often signed by whoever owns the relationship
- Blog author pages — small companies, the author is often the owner
- Contact pages with department breakdowns
ScrapeMaster extracts these into a structured table without CSS selectors — auto-detecting repeating patterns, following links to detail pages, and exporting to CSV, XLSX, JSON, or clipboard for Google Sheets. Extracted data is stored locally in your browser.
The legal line, stated up front: business-role contact information (a named managing director, a role-based email, a business phone) is treated very differently from personal data, but in the EU and UK a named individual is still personal data even in a business context. Stick to publicly published, business-context information; document why you're processing it; honour opt-outs. The compliance section below is not decoration.
Why directories fail at this
Business directories are optimised for coverage, not depth. They'll tell you a company exists, roughly what it does, and how to reach reception. The name of the person who signs off a purchase is not in the data model.
What directories reliably give you: company name, category, address, switchboard number, generic email, sometimes employee-count band.
What they don't: who runs it, who owns the budget, who to actually talk to.
The company's own website is a different kind of source. It's maintained by the company, it's the version they want public, and — crucially — it's structured. A team page is a repeating pattern of name / title / photo / bio, which is exactly the shape an extraction tool handles well.
Page by page
Team, About, and Leadership pages
The highest-yield source and the most consistently structured.
Common URL patterns worth trying directly:
/team /about/team /our-team
/leadership /about/leadership /management
/people /who-we-are /about-us/our-people
/staff /meet-the-team /company/leadership
What's typically on the page: name, job title, photo, short bio, sometimes a direct email, sometimes a LinkedIn link.
The extraction shape: a repeating card or row per person. ScrapeMaster auto-detects this pattern in a couple of seconds and produces a table with sensible column names. Rename the columns, drop the ones you don't want, extract.
Where the depth is: many team pages link each person to a detail page with a longer bio and direct contact details. Enable "follow detail pages" and the extra fields get merged back into the main table. This is the difference between a list of names and a usable contact list — and it's the step most people skip.
Bio text is underrated. It frequently contains tenure ("joined in 2019"), scope ("leads our operations across three sites"), and prior employer. For qualification, that's often more useful than the title.
The footer
Overlooked, and for certain jurisdictions it's the most reliable source of all.
Many countries require specific disclosures on a company website:
- Germany and Austria: the Impressum is legally required and typically names the managing director (Geschäftsführer), registration court, and register number
- UK: company websites must state the registered name, number, and registered office; directors are often listed
- Several EU states: comparable requirements under e-commerce rules
This makes footers and imprint pages a dependable, legally mandated source for who runs the company — far more reliable than a marketing team page that may be out of date.
Legal pages
Privacy policies name a data controller — often with a named Data Protection Officer or privacy contact and a direct email. That's a real person with a real role and a published contact.
Terms and conditions name the contracting entity, which tells you the legal structure — useful when a brand and its operating company differ.
Legal pages are dull and consistently productive. They're also updated more carefully than marketing pages, because someone in legal owns them.
Press and news
Press releases quote executives, and quotes are attributed: name, full title, sometimes a direct media contact at the bottom.
This is the best source for recency. A team page may be two years stale; a press release from March tells you who held that title in March. Extract the archive listing, then follow through to each release.
Careers pages
Job postings leak organisational structure. "Reporting to the Head of Operations" tells you the role exists and sits above the posted one. Some postings name the hiring manager directly.
The pattern of what's being hired also tells you where a company is investing, which is qualification data disguised as recruitment data.
Case studies, testimonials, and blog authors
Case studies are usually signed by the account owner or the relationship lead.
Blog author pages matter for small companies specifically: in a 15-person business, the person writing the blog is often the founder. Author pages typically carry a bio, a title, and social links.
The extraction workflow
Single site
- Open the team page
- Click ScrapeMaster — it opens in the side panel and auto-detects in 2–4 seconds
- Review the detected table, rename columns, remove ones you don't want
- Enable "follow detail pages" if each person links to a profile
- Extract, then export as CSV or copy to clipboard for Google Sheets
Under a minute for a typical team page.
Many sites
The realistic workflow is a loop, not magic: open a site, extract, next. Two things make it much faster:
Saved configurations. ScrapeMaster stores column setup, pagination rules, and detail-follow configuration per domain, so revisiting a site re-applies what you set up. That's most useful for sites you monitor repeatedly.
Consistent columns from the start. Decide your schema before you begin — company, person_name, title, email, source_url, date_captured — and normalise each extraction into it. Merging fifty CSVs with different column names later is the tedious part, and it's entirely avoidable.
A note on honesty: every site is laid out differently, and there's no tool that turns "a list of 500 companies" into "a list of 500 owners" with one click. Auto-detection removes the selector-writing, which is most of the work. It doesn't remove the visiting.
Following detail pages
Worth understanding because it's where most of the value is.
The listing page gives you name and title. The detail page gives you email, phone, full bio, and often direct social links. "Follow detail" opens each item's link in a background tab, extracts your defined fields, and merges them into the main table.
Practical notes: it's slower — you're loading one page per person. Configure a delay if you're doing many. And check the first few results before running the whole set; a mis-mapped field on 200 rows is a re-run.
Data quality
Extraction is the easy part. Usable data is the hard part.
Verify before you use
Cross-check against a second source. A name on a team page plus the same name in a recent press release plus the same name in the footer's registered details is a confident record. A name on a team page alone is a hypothesis.
Check the date. Team pages go stale. Press releases carry dates. Where you can, capture when the page was last modified, and treat anything over 18 months old as unverified.
Watch for role inflation. "Head of" in a 12-person company and "Head of" in a 12,000-person company are different jobs. Cross-reference against company size.
Deduplicate properly
The same person appears as "Jon Smith", "Jonathan Smith", and "J. Smith" across three pages. Normalise on a stable key — email domain plus surname works better than display name.
Record provenance
Store the source URL and capture date on every row. This isn't bureaucracy: under GDPR you may need to tell someone where you got their data, and "a spreadsheet" is not an answer. It's also how you re-verify later without redoing the search.
The compliance part
This section is short because the rules are, and skipping it is how people get in trouble.
Business contact data vs personal data
In the US, business contact information in a commercial context is treated much more permissively. CAN-SPAM governs commercial email and requires accurate headers, a valid physical address, and a working opt-out — it does not require prior consent.
In the EU and UK, a named individual is personal data even at work. "Jane Okafor, Operations Director, jane.okafor@company.com" is personal data. You need a lawful basis — legitimate interests is the usual one for B2B — and that requires you to actually do the balancing test, provide privacy information, and honour objections.
The practical distinction that matters most:
| Lower risk | Higher risk |
|---|---|
info@, sales@, press@ | firstname.lastname@ |
| Switchboard numbers | Direct mobile numbers |
| Registered company details | Inferred personal details |
| Published role and title | Home address, personal social accounts |
| Data the company chose to publish | Data assembled from multiple sources into a profile |
That last row is the one people underrate. Combining several individually-innocuous public facts into a personal profile is a materially different activity from reading a published team page — and regulators treat it that way.
The 2026 environment
Twenty US states now have comprehensive privacy laws in effect. California's Delete Request and Opt-Out Platform (DROP) went live for registered data brokers on August 1, 2026, requiring them to check it at least every 45 days and process consumer deletion requests. Connecticut has enacted its own data broker registration law.
If you sell or share contact data you didn't collect directly from the individuals, you may meet a data broker definition and incur registration obligations. That's a genuine question to ask your counsel, not a hypothetical. Building a list for your own outreach is a different activity from selling one.
Practical rules
- Prefer role-based over personal addresses where either will do
- Never bypass a login to reach data — that changes the legal analysis entirely, and ScrapeMaster doesn't do it: it extracts what's already visible in your browser
- Respect
robots.txtand reasonable rate limits - Keep provenance — source URL and date, per record
- Honour opt-outs immediately, and keep a suppression list
- Don't scrape data you have no use for. Data minimisation is both a legal principle and good hygiene
We cover the wider legal landscape in is web scraping legal, and the tactical side of reaching those contacts in finding company decision-maker contacts.
Tool comparison
| ScrapeMaster | Octoparse / ParseHub | Web Scraper.io | Manual | |
|---|---|---|---|---|
| Setup per site | Auto-detect, 2–4s | Visual builder, minutes | Sitemap config | — |
| Selectors required | No | Sometimes | Yes | — |
| Follows detail pages | Yes | Yes | Yes | — |
| Logged-in pages | Yes (your session) | Varies | Yes | Yes |
| Where data is stored | Locally, in-browser | Often cloud | Local / cloud | Wherever you put it |
| Export | CSV, XLSX, JSON, clipboard | CSV, Excel, API | CSV | — |
| Cost | Free | Paid tiers | Free + paid cloud | Your time |
| Scheduled runs | No | Yes | Cloud tier | No |
The honest positioning: if you need scheduled, unattended, large-scale collection with proxy rotation, a hosted platform is the right category and ScrapeMaster isn't it — it doesn't rotate proxies or fingerprints, and it runs when you run it.
If you need to extract structured data from sites you're browsing, including ones you're logged into, without writing selectors or sending your data to a vendor's cloud, that's what it's for.
Frequently asked questions
Where on a company website are owner and manager details usually found?
Team, About, and Leadership pages first — they're structured as repeating name/title/bio blocks and yield the most. Then the footer and imprint, which in Germany, Austria, and the UK carry legally required company and director details. Then privacy policies (which name a data controller), press releases (which attribute quotes to named executives with dates), and careers pages (which reveal reporting lines).
Can I extract manager contact details without knowing CSS selectors?
Yes. ScrapeMaster auto-detects repeating patterns on a page within a few seconds and proposes named columns. You rename or remove columns and extract — no selectors, no code. For per-person emails and bios, enable "follow detail pages" and it opens each profile link and merges those fields into your table.
Is it legal to extract owner and manager names from company websites?
Extracting publicly published business information is generally lawful in the US, EU, and UK. It gets complicated when the data identifies individuals: in the EU and UK a named person is personal data even in a business context, so you need a lawful basis, you must provide privacy information, and you must honour objections. Avoid bypassing logins, respect robots.txt and rate limits, and prefer role-based addresses where they'll do.
What's the difference between business contact data and personal data?
info@company.com and a switchboard number are organisational data with low risk. jane.okafor@company.com and a direct mobile identify an individual and are personal data under GDPR and UK GDPR even though they're work details. The higher-risk activity is assembling multiple public facts into a personal profile, which regulators treat differently from reading a published team page.
Does extracted data get uploaded anywhere?
No. Extracted records are stored locally in your browser's IndexedDB. The only network request is during auto-detect, when the page's HTML structure — not its content — is sent to the analysis API to suggest columns. Your extracted data is never uploaded.
How do I keep extracted contact lists accurate?
Cross-check every record against a second source on the same site (footer details or a dated press release), store the source URL and capture date on every row, treat anything over about 18 months old as unverified, and normalise names on a stable key like email domain plus surname rather than display name. Re-verify before any campaign.
Can it extract from pages I'm logged into?
Yes, because it works on the page already rendered in your browser using your normal session. It does not bypass paywalls, defeat logins, or solve CAPTCHAs — if you can see it, it can extract it. Note that being logged in usually means you accepted terms of service, which may restrict automated extraction; that's a contractual question worth checking.
Bottom line
Owner and manager details are on the company's own website, and they're on a short list of predictable pages. Team and leadership pages give you the most; footers and imprints give you the most reliable; press releases give you the most current.
The extraction is the easy part — auto-detection removes the selector-writing, and following detail pages is where the contact details actually are. The work that determines whether the list is worth anything is verification, deduplication, and recording where each row came from.
And be deliberate about the line between organisational data and personal profiles. Reading a published team page is one activity; assembling individuals' details from across the web into a profile is another, and 2026's regulatory environment — twenty state privacy laws, California's DROP live since August 1 — treats them differently.
ScrapeMaster is free with no row limits, requires no account, and keeps extracted data in your browser. Pair it with Convert: Web to PDF when you need a dated snapshot of the page a record came from.
Also ours: CineMan AI — it picks what to watch, so you stop scrolling catalogues at 10pm.