Website Tech Stack Detector: CMS, Frameworks, Analytics

What does Website Tech Stack Detector do?

Website Tech Stack Detector shows which technologies any website uses: CMS, ecommerce platform, JavaScript frameworks, analytics and tag managers, CDN, web server, marketing tools and the domain's email provider. Give it a list of domains and it returns one clean record per website, with each technology's name, categories, version (when visible) and how it was detected.

It uses 7,600+ open-source technology fingerprints in the Wappalyzer format and checks each website's response headers, cookies, meta tags, script URLs, HTML and DNS records. It needs no browser and no API key, so it is fast and cheap to run on thousands of domains. On Apify you also get API access, scheduling, webhooks and integrations (Make, Zapier, n8n, Google Sheets, HubSpot).

You are never charged for a failed or empty result

Every result says what it is, and only real results cost anything.

status	Meaning	Charged
ok	A real result	Yes
empty	Worked, but there was nothing to report	No
blocked	The website refused the request (bot protection, HTTP 403 or 429)	No
failed	Could not be done — the error field says why	No
truncated / partial	Cut short by a limit you set; what was delivered is kept	Only the delivered part
skipped	Not attempted, for example after the maximum cost per run was reached	No

If every input fails, the run itself fails with an explanation instead of reporting success with no data.

The Actor also watches the clock: it stops starting new websites two minutes before the run timeout and says in the run summary how many were left. Raise the timeout or split the list if you see that.

Why use Website Tech Stack Detector?

• Lead generation and sales prospecting: find every Shopify store, WordPress site or HubSpot customer in a list of domains, or prospects that use a competitor.
• Competitor research: see which frameworks, analytics and marketing tools competitors run.
• Market research: measure technology adoption across an industry or a country's top websites.
• Agencies and developers: audit client sites before a migration or a pitch.
• Data enrichment: add cms, ecommerce and emailProviders columns to your CRM or spreadsheet.
• AI agents: a quick "what is this website built with?" tool through the Apify API and MCP server.

How to detect a website's technology stack

1. Click Try for free.
2. Paste domains into Websites, one per line (for example apify.com or https://www.example.com).
3. Optionally pick Only these categories (for example CMS and Ecommerce) or turn off Check DNS records.
4. Click Start. Most websites take 1–3 seconds.
5. Open the Output tab. Overview shows one row per website; All technologies shows one row per technology.

Input

Field	Description	Default
Websites	Domains or URLs, one per line. Duplicates and invalid entries are skipped.	–
Check DNS records	Look up MX, TXT, NS and SOA records to find the email provider, DNS host and SaaS tools that verified the domain	on
Only these categories	Keep only technologies from the selected categories	all
Websites in parallel	1–50	10
Timeout per website	Seconds to wait for a website before giving up	20
Extra pages per website	Also check this many other pages on the same site (0–5)	0
Minimum confidence	Drop detections below this confidence (0–100)	0
Security headers report	Report which security headers are set and what each gap means	On

Example input:

{
  "startUrls": ["apify.com", "wordpress.org", "www.allbirds.com"],
  "includeDns": true
}

Output

One item per website. A real example from a run on 15 September 2026 (technology list shortened):

{
  "url": "https://wordpress.org/",
  "finalUrl": "https://wordpress.org/",
  "domain": "wordpress.org",
  "statusCode": 200,
  "technologies": [
    { "name": "Gutenberg", "categories": ["WordPress plugins", "Editors"], "version": "23.9.1", "confidence": 100,
      "website": "https://github.com/WordPress/gutenberg", "detectedBy": ["dom", "scriptSrc"] },
    { "name": "WordPress", "categories": ["CMS", "Blogs"], "version": "7.2", "confidence": 100,
      "website": "https://wordpress.org", "detectedBy": ["dom", "headers", "html", "meta", "scriptSrc"] }
  ],
  "technologyCount": 13,
  "categoriesSummary": { "CMS": ["WordPress"], "Programming languages": ["PHP"], "Web servers": ["Nginx"] },
  "cms": ["WordPress"],
  "ecommerce": [],
  "analytics": [],
  "frameworks": [],
  "server": ["Nginx"],
  "cdn": [],
  "emailProviders": [],
  "dnsServices": [],
  "error": null,
  "scrapedAt": "2026-09-15T11:51:13.335687+00:00"
}

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Output fields

Field	Meaning
url, finalUrl, domain, statusCode	The address you entered, where it redirected to, and the HTTP status
technologies	The website's technology stack: name, categories, version, confidence (0–100), website, detectedBy
securityHeaders	Security header report: score, present, missing with meanings, version disclosure, notes
hosting	ipAddresses, nameservers, cname and the providers those names identify
pagesRead	Which page URLs were actually read for this website
technologyCount, categoriesSummary	Number of technologies and the names grouped by category
cms, ecommerce, analytics, frameworks, server, cdn	Ready-made columns for the most requested categories
emailProviders	Email service from the domain's MX records, for example Google Workspace or Microsoft 365
dnsServices	Services found only in DNS records (domain-verification TXT records, DNS host). They show tools a company has connected its domain to, not the website's code
error	Why a website could not be analyzed (not charged)

detectedBy lists the evidence: headers, cookies, meta, scriptSrc, scripts, html, text, css, url, dom, dns or implied (for example PHP is implied by WordPress).

Security headers, in the same run

Every result carries a securityHeaders report, read from the response we already fetch — so it costs nothing extra and turns a lead-generation run into an audit at the same time:

"securityHeaders": {
  "score": 33,
  "present": ["strict-transport-security", "x-content-type-options"],
  "missing": [
    { "header": "content-security-policy", "meaning": "No policy limiting where scripts may be loaded from." },
    { "header": "referrer-policy", "meaning": "Full URLs may leak to other sites in the Referer header." }
  ],
  "disclosure": { "server": "nginx/1.25.3" },
  "notes": ["Headers name the software and version running the site: server: nginx/1.25.3"]
}

Every missing header says what the gap means, not just its name. Weak settings are called out too — an HSTS max-age under six months, or a CSP that allows unsafe-inline — because a header that is present but toothless would otherwise score as a tick.

Where the site is hosted

hosting names the provider from the site's own DNS records rather than guessing from IP ranges, which go stale and mislabel anything behind a CDN:

"hosting": {
  "ipAddresses": ["23.227.38.33"],
  "nameservers": ["gold.foundationdns.com", "gold.foundationdns.net"],
  "cname": null,
  "providers": ["Cloudflare"]
}

That example is real: Shopify is served by Foundation DNS, Cloudflare's enterprise brand, whose name never contains "cloudflare" — the SOA record is what gives it away.

Looking past the homepage

A homepage is a marketing page. Checkout, analytics and help-desk tools usually live somewhere else. Set Extra pages per website and the Actor also reads up to five other pages on the same site, preferring links it finds on the homepage over guessed paths, and merges what it finds — the strongest evidence wins and a version found on any page is kept. pagesRead lists exactly which pages were used.

It is still one charge per website, however many pages were read.

How much does it cost to detect website technologies?

This Actor uses pay-per-event pricing: you pay a small fee per website analyzed. Websites that can't be loaded, time out or refuse access (HTTP 4xx/5xx) are never charged. See the Pricing tab for the current price per 1,000 websites; platform compute is included. You can set a maximum cost per run and the Actor stops cleanly when it is reached.

Tips

• Enter the homepage. Most technologies are visible there. Enter a specific URL (for example a shop page) to analyze that page instead.
• Bulk lists: keep Websites in parallel at 10–20 for large lists and raise the timeout for slow sites. The default 512 MB of memory analyzes about 40 websites per minute; 1–2 GB is faster for lists of tens of thousands.
• Only need the CMS or shop platform? Select those categories to get smaller results.
• Turn off DNS checks if you only care about the website itself and want slightly faster runs.

Limitations

• No browser: technologies that only appear after JavaScript runs in a browser (JavaScript globals, rendered DOM properties) are not evaluated, and complex CSS selectors in fingerprints are skipped. Server-rendered HTML, scripts, headers and cookies cover most CMS, ecommerce, analytics and marketing tools.
• Blocked websites: some sites refuse automated requests (HTTP 403 or 429) or answer with a bot-protection challenge. They are recognised as soon as the response arrives, reported as blocked with the reason, and are not charged — the Actor does not spend time fingerprinting a challenge page. Sites behind bot protection usually need a browser and residential proxies, which this Actor does not use.
• Versions appear only when a website exposes them.
• DNS services reflect domain-verification records, which can remain after a company stops using a tool.

Related tools

Learn more about the same websites and companies:

• SEO Audit & Metadata Checker: check titles, meta descriptions, canonicals and other on-page SEO for pages or whole sites.
• Broken Link Checker: crawl a website and find broken links, images and redirect chains.
• Company Career Page Jobs Scraper: get all open jobs from company career pages (Greenhouse, Lever, Ashby and more).

FAQ and support

Where do the fingerprints come from? From the open-source webappanalyzer project (a maintained continuation of Wappalyzer's technology data, GPL-3.0), bundled at a fixed version and refreshed periodically.

Does it collect personal data? No. It reads technical signals only (headers, markup, public DNS records) and does not store page text, emails or phone numbers.

Is it legal? It loads one public page per website, like a browser visit, and reads public DNS records. Use the results in line with applicable laws and the websites' terms.

Missing a technology or seeing a wrong result? Open an issue in the Issues tab with the domain. Custom enrichment pipelines are available on request.