Skip to content

Website Industry Classifier

Try for free

Find the industry, category and B2B/B2C business model of any company website, plus a one-line summary, products, customers and a NAICS code. Paste domains or classify the websites of another run's results.

🏷️ Website Industry Classifier

Paste a list of company websites or domains and get each company's industry, category and B2B/B2C business model, plus a one-sentence summary, main products, target customers and a NAICS code. The actor reads each home page (and the about page when the home page says little) and an AI model classifies it, so 20 websites take about 15 seconds and 1,000 about 4 minutes. No login and no API key needed: $2.90 per 1,000 classified websites on the Free plan, less on paid plans.

✨ Why use Website Industry Classifier?

  • 🏢 Industry you can count — every company gets one of about 120 standard B2B industry names, so lead lists can be filtered, grouped and matched to a CRM. Prefer your own categories? Paste them and every website is matched to exactly one of them.
  • 🤝 B2B or B2C — a clear business model per company (B2B, B2C or both), plus isB2B and isB2C flags for quick filters.
  • 📝 More than a label — company name as the site writes it, a more specific category, a one-line summary, up to five products or services, target customers, NAICS code, site language and a confidence score.
  • 🔗 Works on messy lists — plain domains, full links, www. variants and even e-mail addresses (jane@stripe.com) are cleaned up; duplicates are skipped and broken entries get a clear reason.
  • 🧩 Enrich another run — point it at the results of a maps, directory or lead list run and each row comes back with its classification next to the original data.
  • 🌍 Any language — reads websites in any language and writes the results in the language you choose.

🎯 Use cases

  • 🎯 Lead qualification — keep only B2B software companies, or only consumer brands, from a list of thousands of domains.
  • 🧹 CRM cleanup — fill empty or inconsistent industry fields with one standard list or your own.
  • 🗺️ Local business enrichment — classify the websites found by a maps or directory run to build targeted outreach lists.
  • 📊 Market mapping — group a list of startups, suppliers or competitors by industry and niche.
  • 🏦 Due diligence and KYB — see at a glance what a company does, for whom, and under which NAICS code.

🚀 How to use

  1. Paste websites or domains into Websites or domains, one per line (or give the dataset ID of another run's results).
  2. Optional: choose Industry names (built-in list or the model's own words) or paste Your own industry list.
  3. Click Start. Each website becomes one row in the results.
  4. Download the results as JSON, CSV or Excel, or read them through the API.

📥 Input

FieldWhat it does
websitesDomains, links or e-mail addresses, one per line (startUrls, urls and domains are accepted too)
datasetIdClassify the websites in the results of another run
websiteFieldFor datasets: the field that holds the website (found automatically when empty)
maxItemsStop after this many websites (empty = all)
industryListstandard (built-in list, default) or own (the AI model's own words)
industriesYour own industry names; every website is matched to one of them
alwaysReadAboutPageAlso read the about page of every website (it is read anyway when the home page has little text)
outputLanguageLanguage of summaries, categories, products and customers (default English)
modelThe AI model, as author/model (default google/gemini-2.5-flash-lite)
proxyConfigurationProxies used to read the websites (Scrape.ly datacenter by default)
{
  "websites": ["stripe.com", "patagonia.com", "https://www.maersk.com", "deloitte.com", "redcross.org"],
  "industryList": "standard",
  "outputLanguage": "English"
}

📤 Output

FieldDescription
websiteThe website as you gave it
urlThe page that was read (after redirects)
domainDomain without www.
companyNameCompany or brand name as the website writes it
industryBuilt-in industry name, the model's own words, or one of your industries
subIndustryA more specific category or niche
businessModelB2B, B2C or B2B and B2C
isB2B / isB2CQuick filters for the business model
summaryOne sentence: what the company does and for whom
productsAndServicesUp to 5 main products or services
targetCustomersWho the customers are
naicsCode / naicsTitleNAICS 2022 code of the main activity (the model's estimate) and its title
languageMain language of the website
confidenceHow sure the model is of the industry, 0 to 1
basisWhat was read: website, website and about page, or domain name only
pageTitle / metaDescriptionTitle and meta description of the home page
pagesReadLinks of the pages that were read
noteRemarks, for example why only the domain name could be used
errorWhy a website was not classified (these rows are free)

Example rows from a test run:

[
  {
    "website": "maersk.com",
    "url": "https://www.maersk.com/",
    "domain": "maersk.com",
    "companyName": "Maersk",
    "industry": "Logistics & Supply Chain",
    "subIndustry": "Integrated container logistics and supply chain services",
    "businessModel": "B2B",
    "isB2B": true,
    "isB2C": false,
    "summary": "Maersk provides integrated logistics and supply chain solutions for businesses managing shipments and cargo globally.",
    "productsAndServices": ["Inland transport", "Container shipping", "Supply chain management", "Customs brokerage", "Warehousing"],
    "targetCustomers": "Businesses managing shipments and cargo",
    "naicsCode": "488510",
    "naicsTitle": "Customs Brokers and Freight Forwarders",
    "language": "en",
    "confidence": 0.95,
    "basis": "website",
    "pageTitle": "Maersk | Integrated Container Logistics & Supply Chain Services",
    "metaDescription": "Maersk is an integrated logistics company that offers supply chain solutions for managing shipments and cargo. Learn how to register, book and find prices.",
    "pagesRead": ["https://www.maersk.com/"],
    "model": "google/gemini-2.5-flash-lite",
    "note": null,
    "error": null
  },
  {
    "website": "shopify.com",
    "url": "https://www.shopify.com/",
    "domain": "shopify.com",
    "companyName": "Shopify",
    "industry": "Software Development",
    "subIndustry": "E-commerce platform",
    "businessModel": "B2B",
    "isB2B": true,
    "isB2C": false,
    "summary": "Shopify provides an all-in-one commerce platform for businesses to build and grow their online stores, manage operations, and sell across various channels.",
    "productsAndServices": ["E-commerce platform", "Online store builder", "Payment processing", "Shipping tools", "Commerce AI"],
    "targetCustomers": "Businesses and entrepreneurs",
    "naicsCode": "513210",
    "naicsTitle": "Software Publishers",
    "language": "en",
    "confidence": 0.95,
    "basis": "website",
    "pagesRead": ["https://www.shopify.com/"],
    "note": null,
    "error": null
  }
]

A summary of the whole run (counts per industry and business model, websites not classified and why, AI model tokens) is saved as OUTPUT in the run's key-value store.

💰 Pricing

  • $2.90 per 1,000 classified websites on the Free plan; $2.70 on Starter, $2.50 on Scale and $2.10 on Business.
  • Websites that could not be classified (domain does not exist, not a website address, nothing to read) are free.
  • Platform usage is billed on top: AI model tokens, proxy traffic and compute. With the default model that came to about $0.40 per 1,000 classified websites on paid plans and about $0.55 on the Free plan in our test runs (a larger model costs more).
  • Example: 1,000 websites on the Free plan ≈ $2.90 + about $0.55 usage = about $3.45.

❓ FAQ

Is it legal? The actor reads public home and about pages, the same pages any visitor sees. Check the rules that apply to your use of the results.

Do I need a login, an AI API key or a proxy? No. The AI model and Scrape.ly proxies are built in; usage is billed to your Scrape.ly account.

How many websites can I classify? As many as you like: there is no limit on the number of websites. Big lists are processed in parallel and each result is saved as soon as its website is done.

What if a website refuses access? Some sites block automated visits. For companies the AI model knows well, the actor still classifies them from the domain name and says so in basis (domain name only) and note, with a lower confidence. Unknown companies whose site cannot be read come back with a reason in error and are free.

How accurate is it? The industry and business model are taken from what the site itself says, and confidence shows how plainly it says it. Use your own industry list when you need exact CRM values, and a larger model for hard cases. The NAICS code is the model's estimate.

Can I schedule it or use the API? Yes. Schedule runs, call it through the Scrape.ly API, or chain it after another run with its dataset ID.

⚠️ Limits

  • Only the home page (and one about page) is read; the first 6,000 characters of the home page and 4,000 of the about page go to the AI model.
  • Websites that show nothing without JavaScript, or that refuse automated visits, are classified from the domain name only when the company is well known; otherwise they are not classified.
  • Built-in industry names are in English; summaries, categories and products follow Language of the results.
  • One industry per company: groups with many businesses get the industry of their main activity.

Related actors

//From the blog\\

Latest articles

All articles
G2
4.7/5
Excellent
Trustpilot

The internet is full of data.
Start owning yours.

Scrape.ly gives you the infrastructure, the tooling, and the team to extract, process, and operationalize web data at any scale. Setup takes minutes.

Get Started FreeChat with Support
No credit card. No setup calls.