Target company URL research is the process of taking a company name and finding the official website that belongs to it, then confirming that the website really is the right company.
Finding one is trivial. You search the name, you click the site, you are done in nine seconds. The work only becomes work at volume. A list of 400 company names with an empty website column is somewhere between a long afternoon and two days, and the tedium is what makes it error-prone. People stop checking around row sixty.
The part almost everyone underestimates is the second half. Finding a plausible domain is easy and mostly automatable. Confirming that the domain belongs to the company you actually meant is neither, and it is where the value sits. A wrong domain does not announce itself. It sits in the CRM looking exactly like a right one, and everything built on top of it inherits the mistake.
This guide covers the whole process: how name-to-domain matching works, what a confidence score does and does not tell you, the specific checks that confirm ownership, how to run this on a list of a thousand names without checking a thousand websites by hand, and which free data sources are still standing in 2026.
What “target company” actually means here
A target company is any business you are researching with intent. A prospect on a sales list, an employer you want to apply to, a competitor you are mapping, a supplier you are vetting, an acquisition candidate.
The term is standard B2B vocabulary and has nothing to do with Target Corporation, the American retailer. If you arrived here looking for that company, its sites are target.com and corporate.target.com, and you want a different page.
URL or domain: the distinction matters more than it sounds
Most people use these interchangeably. For research purposes they are not the same thing, and storing the wrong one causes problems later.
A domain is the root address: stripe.com. A URL is a specific address on that domain: https://stripe.com/pricing.
For a company record you almost always want the bare domain. It is the stable identifier, it is what email addresses hang off, it is what CRMs deduplicate against, and it is what SEO and enrichment tools expect as input. A URL pointing at a marketing landing page or a regional subfolder will break all of that quietly.
Store the domain as the primary field. Store specific URLs, careers pages and regional sites in their own fields if you need them.
The five-stage process
Every reliable workflow runs the same sequence, whether you are doing it by hand for ten companies or with a tool for ten thousand.
Company name → clean name → candidate domains → verification → stored record
The middle stage produces guesses. The fourth stage turns guesses into data. Skipping the fourth stage is the single most common failure in this whole discipline, and it is why so many prospect lists contain confidently wrong websites.
Stage 1: clean the input
Bad input is responsible for most misses. Before you look anything up:
Strip the legal suffix. Companies register as one thing and trade as another. Bolt Technology OÜ matches badly. Bolt matches well. Remove Inc, Incorporated, Ltd, Limited, LLC, LLP, PLC, GmbH, SA, BV, AB, Pty Ltd and Pvt Ltd.
Do not strip blindly, though. Some of those words are genuinely part of the brand, and a few company names are almost entirely suffix. Check that removal leaves something meaningful behind.
Use the trading name. Alphabet Inc. and Google are the same organisation with very different match behaviour. Whichever name the market uses is the one databases have indexed.
Fix the obvious mess. Trailing commas, double spaces, two companies crammed onto one line, blank rows, mixed casing, stray quote marks from a spreadsheet export. These cause more failures than any matching algorithm.
Deduplicate. The same company appears twice under slightly different spellings more often than you would think, particularly in lists assembled from multiple sources.
Keep the context columns. If your list has country, city or industry, do not discard them. They are the only thing that will resolve ambiguous names later, and they are far more valuable at this stage than any scoring refinement.
Stage 2: find candidate domains
You have three broad routes, and the honest answer is that they trade off cost against coverage.
Manual search. Still the best method for hard cases. The trick is to add context to the query rather than searching the bare name: "Pioneer Solutions" software India official website beats Pioneer Solutions by a wide margin. Ignore the directory results, the aggregators and the LinkedIn page, and look for the company’s own domain.
Free APIs. Clearbit’s Autocomplete endpoint is the one that still works without payment. Clearbit’s own help documentation describes the Autocomplete API as a completely free product usable without an account or an API key, and it accepts a partial company name and returns matching company names with their primary domains.
Two things to know before you build anything on it. First, Clearbit documents a limit of 600 requests per minute. Second, and more important, the surrounding product is in wind-down. Free Clearbit accounts, the free platform, the TAM calculator, Clearbit Connect and the free implementations of the Name to Domain and Risk APIs were all retired on 30 April 2025, and the free Logo API followed on 8 December 2025. Autocomplete is the last free endpoint left on a product HubSpot is steadily folding into Breeze Intelligence. Use it, but do not architect around the assumption that it will be there in a year.
Wikidata. A genuinely free, genuinely stable second source. Wikidata’s P856 property records the official website of an entity, so you can resolve a company entity and read its official domain straight off. The limitation is coverage: Wikidata is well populated for large and well-known companies and close to empty for small and regional ones, which means it is useful as a cross-check rather than as a fix for the cases where commercial databases fail.
Paid providers. If accuracy at volume is the job, this is what you are actually buying. Coverage of small and regional companies is the thing free sources cannot give you, and no amount of clever scoring substitutes for a database that contains the company.
One route to avoid: automated scraping of search engine results. It is fragile, it breaks without warning, it creates terms-of-service exposure, and the first organic result is frequently a directory rather than the company.
Stage 3: read the confidence score correctly
Most lookup tools return a score or a High / Medium / Low label. Understanding what that number measures is the difference between using these tools well and being misled by them.
Almost every confidence score is a string similarity measure. It compares the name you submitted against the name stored in the database. That is all. It tells you the two pieces of text resemble each other. It does not tell you the domain belongs to the organisation you had in mind.
This is worth sitting with, because it produces a specific and dangerous failure mode: a perfect score on completely the wrong company. Submit Mercury and you can get a flawless name match to a bank, an insurer, a defence contractor or a car brand that no longer exists. The score will be high for all of them, because the string matched perfectly every time.
Here is how to actually treat the bands:
| Label | What it means | What to do |
|---|---|---|
| High | Near-identical name match | Likely correct. Spot check a sample, and always check it if the name is short or generic |
| Medium | Partial match, often a plural or a word difference | Open every one before use |
| Low | Names differ substantially | Treat as a lead to investigate, not an answer |
| No match | Not in the database | Manual search, usually faster than fighting the tool |
The practical rule: confidence scores are for triage, not for truth. They tell you where to spend your review time. They earn their keep by letting you skip verification on the boring 70% so you can spend real attention on the 30% that might be wrong.
Stage 4: verify ownership
This is the stage that separates a usable list from a plausible one. Open the candidate domain and look for identity evidence.
Does the name on the site match?
Small variations are normal and fine. ABC Technology and ABC Technology Ltd. are the same company. A different word entirely is not a variation.
Does the About page support it?
This is the highest-value single check. Founding story, headquarters, markets served, leadership. A real About page will confirm or destroy a match in about fifteen seconds.
Does the industry match?
The most common wrong-company signature is a perfect name match on a business that sells something else entirely. You are looking for a logistics firm and the site sells dental software. The name matched. The company did not.
Does the location make sense?
Check the contact page against whatever country or city you have on the list. Geography resolves same-name collisions faster than anything else.
Do the legal pages agree?
Privacy policy, terms of service and copyright footers usually name the registered legal entity, which is exactly the information you need when a site trades under a shortened brand name. This is the most under-used check on the list and often the most decisive one.
Does the company link to this domain from its own profiles?
A company’s verified LinkedIn or X profile linking to the same domain is strong independent confirmation. This is a good final check on accounts that matter.
Where does it redirect?
A redirect is information, not noise. It usually means a rebrand, an acquisition or a domain migration, and you want to record the destination rather than the old address. Notion moving from notion.so to notion.com in June 2026 invalidated a lot of stored records that had looked fine for years.
For high-value accounts, use two or three independent signals before you mark something verified. For a bulk market-research list, the About page check alone is a reasonable bar.
Stage 5: store it properly
The field you store determines how much of this work you will repeat in six months. A minimum viable company record:
| Field | Why it exists |
|---|---|
| Name as submitted | Lets you trace back to the original source list |
| Matched name | Shows what the database actually returned |
| Primary domain | The stable identifier everything else keys off |
| Regional URL | Prevents losing market-specific sites |
| Country | Resolves future ambiguity |
| Confidence score | Triage signal for re-review |
| Verification status | Verified / Needs review / No match / Rebranded / Inactive |
| Last verified date | Tells you when to look again |
| Source | Which database or method produced this |
Verification status and last-verified date are the two fields teams skip and then wish they had. Without them you cannot tell a checked record from an unchecked one, which means the next person re-does everything or trusts everything. Both outcomes are bad.
Why name matching goes wrong
Understanding the specific failure modes lets you predict which rows in your list will need attention before you run anything.
Names are not unique identifiers: This is the root cause of almost everything below. Nothing stops two unrelated companies in two countries from registering the same trading name, and thousands have.
Short and generic names are the worst offenders: Bolt, Atlas, Apex, Mercury, Summit, Pioneer, Vertex, Nova. Each is used by multiple substantial businesses across different industries. Any lookup on these is a coin flip dressed up with a confidence score. Flag them manually before you start.
Parent, subsidiary and brand get conflated: You search a group name and get the operating company, or you search a product brand and get the parent. Both are defensible answers and only one is the one you wanted. Record the corporate relationship when it matters.
Rebrands and acquisitions decay your data silently: Nothing in your CRM updates itself when a company changes its domain. The record just becomes wrong.
Regional domains fragment the answer: A global business may run example.com, example.co.uk, example.in, in.example.com and example.com/in/ simultaneously. There is no single correct answer, only the right answer for your purpose. Corporate research wants the global domain. Local sales and recruiting usually want the regional one. Store both.
Small companies are simply absent: Local businesses, very new startups and companies with unusual transliterations frequently do not exist in any commercial database. The tool is not broken. The data is not there, and a manual search is faster than retrying.
Lookalike domains exist deliberately: Some domains are built to resemble a real company’s. Extra words like “official” or “login” in the domain, a spelling one character off the real brand, a very recent registration date, or a country extension that makes no sense for that business are all worth a second look before you click or import.
Running this on a real list
The approach should change with volume, and treating every row with equal care is the mistake that makes bulk research feel impossible.
Under 25 companies. Do it by hand. You will be finished before you have finished configuring a tool, and every row gets properly verified.
25 to 200 companies. Automated lookup, then manual review of everything that is not a clean high-confidence match. This is the sweet spot for free tools.
200 to several thousand. Now you need triage, because you cannot verify every row and you should stop pretending otherwise. Split the output three ways:
Group 1, high confidence with a distinctive name. Sample-check ten percent. Accept the rest.
Group 2, medium or low confidence, or any short or generic name regardless of score. Manual verification, every row. This is where your reviewing hours go.
Group 3, no match. Manual search, or accept as unresolved and move on. Do not let these consume the time that Group 2 needs.
The whole point of the confidence score is to build these three piles. Fifteen percent of a list usually needs real human attention, and the reason bulk URL research fails is that teams either check nothing or try to check everything.
Above that, and recurring. Budget for a paid provider. Coverage of small and regional companies is the constraint, and it is the one thing free sources genuinely cannot solve.
How often to recheck
There is no universal interval, but there are clear triggers. Recheck a domain when emails to it start bouncing, when the site stops resolving, when the company announces a rebrand or acquisition, when your CRM data conflicts with what the website says, or when a high-value account has not been touched in a year.
A last-verified date makes all of this a filter instead of a project.
What a URL gets you, and what it does not
A verified domain is the entry point to company research, not the research itself. Setting that expectation prevents a lot of disappointment with free tools.
| Information | Available from the website? |
|---|---|
| Official domain | Yes |
| Legal or registered name | Usually, in the legal pages |
| Products and services | Usually |
| Headquarters and locations | Usually |
| Contact route | Usually |
| Careers page | Often |
| Leadership | Sometimes |
| Pricing | Sometimes |
| Headcount | Rarely stated, and rarely accurate when it is |
| Revenue | Almost never, unless the company is listed |
Headcount, revenue and verified contact details are enrichment data. They come from providers who compile them, not from a name-to-domain lookup. Any tool suggesting otherwise is overselling.
Who this work is for
Sales and SDR teams need the domain before qualification, personalisation or contact research can start. It is the first field in the workflow and everything downstream depends on it.
Recruiters and job seekers use it to reach the official careers page rather than a third-party board. Note that most companies hand off to an external applicant tracking system, so the careers link frequently leaves the company domain. Start on the official site and follow its link rather than guessing at URL patterns, since companies split roughly evenly between /careers, /jobs, /join-us and a subdomain.
SEO and competitive teams need verified competitor domains as input to backlink, keyword gap and content analysis. A wrong domain here does not produce a small error, it produces an entire analysis of an unrelated business.
Market researchers turn a name list into an industry map, working from first-party sources rather than aggregator summaries.
RevOps and CRM owners use domain verification for deduplication and data hygiene, because domain is the field accounts should be matched on and it is frequently the field that is missing or wrong.
A note on doing this responsibly
Company website research uses publicly published business information, which is about as uncontroversial as data work gets. The care is needed at the next step. Once you start attaching named individuals and contact details to those company records, you are handling personal data, and the applicable rules vary by jurisdiction and by how you intend to make contact. Collect what the work requires, hold it only as long as it is useful, and know which rules apply where you operate.
FAQs
What is target company URL research?
The process of identifying the official website or domain belonging to a company you are researching, and confirming the site genuinely belongs to that organisation. It is standard groundwork for sales prospecting, recruiting, SEO competitor analysis and CRM data hygiene.
How do I find a company’s official website?
Search the company name together with its country, industry or “official website” rather than the bare name, then verify the result against the site’s About page, contact details, products and legal pages before you record it.
Can I look up company URLs in bulk?
Yes. Lookup services accept a list of names and return likely domains with a match score. Treat the output as candidates and review anything that is not a clean high-confidence match on a distinctive name.
What does a confidence score actually measure?
In almost every tool, the similarity between the name you submitted and the name stored in the database. It is a text comparison, not proof of ownership, which is why a short name like Bolt or Mercury can score highly on entirely the wrong company.
Is Clearbit’s free API still available in 2026?
The Autocomplete endpoint is, and Clearbit documents it as free and usable without an account or API key. The free Name to Domain API, free accounts and the free platform were retired on 30 April 2025, and the Logo API on 8 December 2025. Autocomplete is the last free endpoint on a product being folded into HubSpot’s Breeze Intelligence, so treat it as usable today rather than dependable long term.
Why did a company not come back in my results?
Free and commercial databases both have gaps, concentrated in very small businesses, recently founded companies and unusual spellings or transliterations. Try the short trading name or the English form of an international brand, then search manually.
Two companies share a name. What now?
Add context to the lookup: country, city, industry, product or parent company. Then verify against the site’s location and what it actually sells. Industry mismatch is the fastest way to spot the wrong company behind a perfect name match.
Should I store the global domain or the regional one?
Both, in separate fields. Corporate research wants the primary global domain. Local sales, recruiting and SEO usually want the country-specific site. Storing only one loses information you will want later.
What happens when a company changes domain?
Old records become wrong without any signal. Check where redirects lead, establish whether it was a rebrand, an acquisition or just a migration, update the primary domain and keep the previous one for history.
Can AI do this automatically?
AI handles the mechanical parts well: cleaning names, deduplicating, comparing similar names, flagging uncertain matches, summarising a site once you have it. It is not reliable as the final authority on which company a domain belongs to, because it will confidently associate similar names and may be working from outdated information. Verify against the live first-party site for anything that matters.
Is a free tool good enough?
For turning a few hundred recognisable company names into domains, yes. For coverage of small and regional companies, or for headcount, revenue and verified contacts, no. Those need a paid provider, and the gap is data coverage rather than feature polish.
In short about- Target Company URL Research
The finding is the easy half. Any reasonable database plus clean input gets you most of the way on well-known companies in seconds.
The verification is the half that determines whether the output is data or decoration, and it is the half that free tools cannot do for you. Confidence scores triage, they do not confirm. Short names lie convincingly. Redirects mean something. First-party pages settle arguments.
Build the habit of recording what you verified and when, and this stops being a recurring project and becomes a field that stays true.