Tech Stack Detector (Dataset, CSV or Sheet)

You have a list of companies. Which of them run Shopify? Who is on HubSpot, who is on WordPress, and whose email lives on Google Workspace?

Point this Actor at the list you already have, an Apify dataset, a CSV, an Excel file or a Google Sheet, and it reads each company's homepage and DNS records and hands back every original column untouched with the technology columns added alongside: the e-commerce platform, the CMS, analytics, marketing automation, payment providers, hosting, email hosting and more, out of 7,600+ technologies. No copying a domain column out, no matching results back to company names by hand.

What it returns

For every row you get:

• technologyNames: the products the site runs, as one readable line, for example Shopify, Microsoft 365, Apple Pay, PayPal, Cloudflare, Google Tag Manager
• Ready-made columns you can filter a spreadsheet on: ecommercePlatform, cms, analytics, marketingAutomation, crm, paymentProcessors, hosting
• Email and DNS columns: emailHosting (Google Workspace, Microsoft 365, Zoho and others, proven by the domain's MX record), emailSending (SendGrid, Mailgun, Amazon SES and others, from its SPF record), dnsProvider, and dnsVerifiedServices
• technologies: the full list, each with its categories, version where the site reveals one, a confidence score, and the evidence it was found from
• pageTitle and pageSizeKb: what page was actually analysed, so you can see at a glance that it was the real homepage and not an error page
• techStatus: ok, no_technologies_found, blocked, unreachable, not_html, disallowed_by_robots or invalid_url

Web standards such as HSTS, HTTP/3 and Open Graph are kept in the full list but left out of technologyNames and the ready-made columns, because they tell a sales team nothing.

Find every Shopify store in your list

Put Shopify in Technologies to look for and set Rows to keep to Only rows matching the filter. You get back only the matching companies, with all their original columns. It works for anything in the fingerprint set: WooCommerce, HubSpot, Klaviyo, Salesforce, Google Workspace, Microsoft 365, Webflow. You can also filter by whole categories such as Ecommerce or Marketing automation, and ask for sites that use any or all of what you listed. Misspell a name and the run tells you what you probably meant.

What makes it different from other tech stack detectors

• It takes your whole list. None of the ten most-used tech stack detectors on the Store accepts a dataset, file or sheet: each takes a typed-in list of URLs and hands back a bare list you then have to re-join to your own data. This one reads a dataset, a file or a Google Sheet directly and keeps every column you had.
• It reads DNS as well as the page. That is how it can tell you a company's email runs on Microsoft 365 while its shop runs on Shopify, and which services the company has verified its domain with.
• It says how it knows. Every technology carries its evidence (a script source, a cookie, a meta tag, a DNS record) and whether it was seen on the website itself or only in DNS.
• It catches tools installed by a loader snippet. Tag managers, email-capture tools and chat widgets are usually injected by a small inline script. The script address inside that snippet is picked up and counted, at a reduced confidence of 75 so you can tell it apart.
• You never pay for a site it could not read. Blocked, unreachable, non-HTML and robots-disallowed sites are free.

Example

A real row from the default example input:

{
  "company": "Allbirds",
  "website": "allbirds.com",
  "techStatus": "ok",
  "pageTitle": "Allbirds: Comfortable, Sustainable Shoes & Apparel",
  "technologyNames": "Shopify, Microsoft 365, Apple Pay, PayPal, Cloudflare, Google Tag Manager, Swiper, Attentive",
  "ecommercePlatform": "Shopify",
  "paymentProcessors": "Apple Pay, PayPal",
  "emailHosting": "Microsoft 365",
  "dnsVerifiedServices": "Cursor, DocuSign, Miro",
  "technologies": [
    { "name": "Shopify", "categories": ["Ecommerce"], "confidence": 100, "detectedOn": "website", "evidence": ["cookie", "meta", "script-loader", "script-src"] },
    { "name": "Microsoft 365", "categories": ["Webmail", "Email"], "confidence": 100, "detectedOn": "dns", "evidence": ["dns-mx", "dns-txt"] },
    { "name": "Attentive", "categories": ["Personalisation"], "confidence": 75, "detectedOn": "website", "evidence": ["script-loader"] }
  ],
  "fetchedVia": "direct",
  "pageRenderedWithJavascript": false
}

The run summary also lists the most common technologies across your whole list, which doubles as a quick market map: how many of your prospects are on Shopify versus WooCommerce, or on HubSpot versus Klaviyo.

How each site is checked

1. robots.txt is read first and honoured by default: a site that disallows its homepage for automated agents is skipped and not charged.
2. The homepage is fetched once, exactly as the server sends it, and matched against the fingerprint set: script sources, inline scripts, HTML, meta tags, cookies, response headers, page structure and the URL itself.
3. DNS records (MX, TXT, NS, SOA, and the web host's CNAME) are looked up, which adds email hosting, email sending, DNS provider and verified services at no extra charge. You can switch this off.

The page's JavaScript is not run, which is what keeps this fast and cheap, and every row says so in pageRenderedWithJavascript: false. The trade-off is honest and worth knowing: a tool that only appears after scripts run can be missed, so a missing technology is not proof that a site does not use it.

Blocked sites. Some large brands refuse automated requests. Those rows come back as blocked and cost nothing. If your list is heavy on big brands you can switch on Retry blocked sites through a residential proxy: a blocked site is retried once through Apify's residential proxy, and only a retry that succeeds is charged. This helps against sites that block cloud-server addresses; sites with advanced bot protection, which is most of the biggest brands, will usually still refuse. Most small and mid-sized business websites are read directly with no trouble.

Pricing

• $0.005 per website analysed
• Nothing for a site that could not be reached, blocked the request, was not an HTML page, or was disallowed by its own robots.txt
• $0.01 per CSV or Excel export, $0.02 per webhook delivery
• $0.005 extra per blocked site successfully read through the residential proxy, only if you switch that option on

So 1,000 websites cost about $5, and 100 cost about 50 cents. DNS lookups, the filter, the summary and the market map are included.

Store discount tiers are on from day one, so Bronze, Silver and Gold plan customers pay 10, 20 and 30 percent less.

FAQ

Is this like Wappalyzer or BuiltWith? It does the same kind of job, and it is not affiliated with either. The difference is the input and the output: you give it your list as it is, a dataset, file or sheet, and you get that same list back with the technology columns added, plus email hosting from DNS.

It missed a tool I know they use. Why? Most likely that tool is only loaded after the page's JavaScript runs, for example a chat widget or an A/B testing script pulled in by a tag manager. This Actor reads the page as the server sends it, which is fast and cheap but cannot see those. Tools installed by an inline loader snippet are still caught.

Why does it say Microsoft 365 when their website runs on Shopify? Those are two different things. ecommercePlatform is what the website runs on; emailHosting is where the company's email lives, read from the domain's MX record. Both are useful and they are reported separately.

What are "DNS-verified services"? Companies add a DNS record to prove they own their domain to services such as Zoom, DocuSign, Atlassian or Salesforce. Seeing one means the company has an account there. It is kept in its own column because it is about the company, not about what the website runs on.

Why is a technology's confidence 75? It was found only as a script address inside an inline loader snippet, which is strong evidence but not proof the tool actually runs. Set a minimum confidence if you only want certain detections.

Where does the fingerprint data come from? From the open-source webappanalyzer fingerprint set, pinned to a known version so every run is reproducible; the exact source version is recorded in each run's summary. One over-broad upstream rule is corrected: an Apple domain-verification record is not treated as proof of iCloud Mail.

Can an AI agent run it? Yes. It uses pay-per-event pricing with limited permissions, so it can be called through Apify's MCP server and paid for with x402.

The rest of the toolkit

The natural pair is Website Contact Finder: run both on the same list, or chain them with the Pipeline Runner, to get every Shopify store in your list together with its business inbox. Then Email List Cleaner & Validator checks the addresses and Phone Number Validator & Cleaner checks the numbers.

For the data itself: Dataset Cleaner & Exporter, Filter & Transform, Join & Merge, Aggregate, Group By & Pivot, Diff & Change Detector, AI Enrich, Charts & Report, to Postgres, Supabase & MySQL, to REST API, and Actor Pipeline Runner to chain them in one call.

---

If this saved you opening a few hundred websites to check what they run on, a review on the Store page helps a lot. If something looks wrong, open an issue on the Issues tab and I will answer personally.