Website Technology Detector with Evidence

Turn a list of website URLs into structured technology observations. Identify supported CMS, ecommerce platforms, analytics tags, marketing tools and infrastructure, with a signal explaining each match.

Use the results to qualify company lists, find stores using Shopify or WooCommerce, or group websites by the tools visible on their pages. This Actor inspects the HTML and response headers from one public page per URL. It does not execute JavaScript or load third-party assets.

Start with two websites

{
  "urls": ["https://apify.com", "https://wordpress.org"],
  "maxConcurrency": 3,
  "timeoutSecs": 20
}

Add URLs, click Start, then export the dataset as JSON, CSV or Excel. JSON retains the full evidence objects. Use technologyNames for spreadsheet filters. You can also run the Actor through Apify's API, integrations and schedules.

What you receive

Each dataset row contains:

• url and finalUrl: requested and final addresses after redirects.
• status: ok, blocked, http_error or error.
• title, httpStatus, scannedAt and durationMs.
• technologyNames and technologyCount: convenient fields for filtering.
• technologies: names, categories and the source/signal for each observation.
• redirects, error, method and limitation where applicable.

For example, an observed Next.js script path produces this technology entry:

{
  "name": "Next.js",
  "category": "Web framework",
  "evidence": [
    { "source": "assets", "signal": "/_next/ asset" }
  ]
}

The evidence describes the matched marker rather than returning cookies, complete response headers or tracking IDs. Framework implications are explicit: a React entry derived from Next.js has inferredFrom: "Next.js" and an empty evidence array. It is not presented as a separate direct observation.

The key-value store record SUMMARY reports requested, delivered, successful and failed counts, plus whether the run reached its charge limit. If the charge limit stops a run, unprocessed URLs do not produce rows.

Price and charge behavior

The launch price is $0.001 per successfully inspected HTML page ($1 per 1,000) under the website-scanned event. A successful inspection with zero supported technologies found is still one billable observation. The price is per page, not per technology.

Blocked pages, HTTP errors, timeouts, invalid URLs and non-HTML responses do not trigger the inspection event. There is no Actor start fee or separate charge per dataset row. The current Pricing tab is authoritative for the active rate.

Duplicate input strings are trimmed and processed once within a run. Distinct URLs, including different paths on the same domain, count separately. Redirect aliases are not deduplicated across inputs. Running the same list again creates new inspections and charges.

Complete example result

This is a complete row from our WordPress.org validation run on 6 September 2026. Website signals can change; the example demonstrates the output format.

{
  "url": "https://wordpress.org/",
  "status": "ok",
  "scannedAt": "2026-09-06T21:33:01.018Z",
  "technologyNames": [
    "Google Tag Manager",
    "Nginx",
    "WordPress"
  ],
  "technologyCount": 3,
  "technologies": [
    {
      "name": "Google Tag Manager",
      "category": "Tag manager",
      "evidence": [
        {
          "source": "assets",
          "signal": "Google Tag Manager gtm.js script"
        }
      ]
    },
    {
      "name": "Nginx",
      "category": "Web server",
      "evidence": [
        {
          "source": "headers",
          "signal": "Nginx server header"
        }
      ]
    },
    {
      "name": "WordPress",
      "category": "CMS",
      "evidence": [
        {
          "source": "generator",
          "signal": "WordPress generator"
        },
        {
          "source": "assets",
          "signal": "/wp-content/ or /wp-includes/ asset"
        }
      ]
    }
  ],
  "error": null,
  "finalUrl": "https://wordpress.org/",
  "httpStatus": 200,
  "redirects": [],
  "title": "Blog Tool, Publishing Platform, and CMS – WordPress.org",
  "method": "http-html-and-response-headers",
  "limitation": "Only the returned HTML and headers were inspected. No JavaScript execution. An absent detection does not prove absence.",
  "durationMs": 388
}

Coverage and limits

The initial rules cover 81 named technologies. Detection requires a supported visible marker; coverage of a product does not guarantee detection on every installation.

• CMS and stores include WordPress, Drupal, Joomla, Ghost, Hugo, Jekyll, Shopify, WooCommerce, Magento, BigCommerce, PrestaShop and Shopware.
• Builders and frameworks include Wix, Squarespace, Webflow, Framer, Duda, Elementor, Divi, Next.js, Nuxt, Astro, Gatsby, SvelteKit and Angular.
• Analytics and marketing include Google Analytics, Google Tag Manager, Plausible, Matomo, Clarity, Hotjar, Mixpanel, Amplitude, Segment, PostHog, HubSpot, Klaviyo, Mailchimp and several advertising tags.
• Other rules cover support widgets, payment scripts, consent tools, CDNs, hosting headers, web servers and observability scripts.

Only the returned HTML and headers are inspected. Technologies loaded after consent, user interaction or JavaScript execution may be missed. Compiled or renamed assets, removed metadata and hidden server components may also be missed. An absent match does not establish that a technology is absent. A matched script can be embedded but inactive. This is an observed-signal report, not a verified inventory of the website's backend.

Limits are 500 unique URLs per run, 1 to 5 concurrent requests, 5 to 40 seconds per URL including redirects, at most 5 redirects, and 3 MB of decompressed HTML per page. HTML is decoded as UTF-8. Only public HTTP(S) destinations on ports 80 and 443 are accepted. Private networks, credentialed URLs, login sessions, PDF files and browser challenges are unsupported. No software versions or vulnerability assessments are produced.

Keep runs small initially. Review error rows before increasing batch size. When inspecting several pages on the same host, use concurrency 1. No residential proxy or separate third-party API subscription is required.

Support

Open an issue from this Actor's Issues tab with a public example URL, the run ID and the missing or incorrect technology. Avoid including passwords or confidential URLs. Technology names belong to their respective owners; this Actor is independent of those products.