Which CMS, which pixel, which stack.
outsend opens each prospect's website, reads the HTML, inspects headers, identifies JS signatures, and instantly tells you which CMS it runs on, which ad pixel it carries, and which CRM or analytics tool it has installed. Tech targeting as precise as a rep who visited every site — at 0.8s per record.
You sell a solution that integrates with WooCommerce, Shopify, or Magento. You want to reach only the 280 businesses actually running one of those three platforms — not the thousands of others built on Wix or Squarespace. Without a tool, you visit each site, inspect the source code, and draw conclusions. For 200 companies, that's a full day wasted opening tabs and reading HTML.
outsend.xyz's tech stack detection automates that work: for every domain you feed it, the tool identifies the main technologies in use (CMS, e-commerce platform, analytics, hosting provider, ad pixel, marketing CRM, support platform…) and returns them as filterable columns on your list. Then you target by stack, not by gut feeling.
Eight technology families detected
What the tool recognizes from a website's HTML signatures, JS, and HTTP headers.
CMS / Front-end
The engine serving the pages.
E-commerce
The online selling platform.
Analytics
Visitor behavior tracking.
Ad pixels
Retargeting and conversion tags.
Hosting / CDN
The infrastructure and acceleration layer.
Payment
The integrated payment processor.
CRM / Marketing
The contact and email management tool.
Chat / Support
The live chat widget.
Two analysis paths, merged
The tool combines two signals to eliminate false positives: static HTML parsing + active JavaScript signal scanning. When one confirms the other, the detection is solid.
HTML + header parsing
Reads the HTML served by the site: <meta> tags, <link> tags, theme-specific CSS classes, external asset URLs, HTTP server headers. Stable structural signals.
meta name="generator"
→ WordPress 6.4
link href="/wp-content/plugins/
woocommerce/..."
→ WooCommerce
Active JavaScript signals
Detects calls to fbq(), gtag(), _paq.push, Didomi.…, hbspt.…, Intercom(), OneTrustActive, Tawk_API. Catches technologies that load their scripts dynamically (SPAs, consent-before-load).
window.fbq → Meta Pixel
window.gtag → Google Ads
_paq.push → Matomo
hbspt.cta → HubSpot CRM
Three use cases where stack detection makes the difference
Filter a list of 1,000 e-commerce sites
You sell a WooCommerce integration. You only want to reach the 280 sites actually running WooCommerce. Without detection, half your emails go to prospects on Shopify or PrestaShop who cannot use your solution.
Tech audit of your 20 competitors
A marketing team wants to know which analytics and tracking tools their direct competitors use. Automated detection returns a full technical profile in minutes, rather than investigating site by site.
Verify the "cloud-native" claim
A VC auditing a startup wants to verify that the "cloud-native infrastructure" claim matches technical reality. Detection provides a fast, objective snapshot that can be cross-referenced against what the startup declares in its pitch.
The market: export-based pricing, mostly in USD
Wappalyzer
Free browser extension for one-by-one lookups. Bulk export requires a paid API plan, priced for the US market.
BuiltWith
Larger historical database (millions of sites). Ideal for macro research; more expensive for ongoing qualification.
SimilarTech
International scope with useful time-series tracking. Not optimized for single-country prospecting workflows.
outsend
Detection built into the pipeline (extraction → tech stack → filtering). Apply for access, no credit-based export.
⚠ What tech stack detection does not do
- No visibility into server-side infrastructure (exact backend language, database, k8s vs VM) — the tool only sees what is served client-side
- No "technical quality" scoring of the site — it lists technologies, it does not evaluate them
- No time-series history ("has been using WordPress for X months") — for that, BuiltWith remains the best option
- No exact version detection when intentionally hidden (e.g.,
Server:header suppressed)
Filter your list by stack
Detection built into the outsend pipeline. Extraction → tech stack → filtering in a few clicks, free alpha on application.
Request free alpha access →Frequently asked questions
What technologies are detected exactly?
Eight families: CMS, e-commerce, analytics, ad pixels, hosting/CDN, payment, marketing CRM, chat/support. Each family covers established players (Wappalyzer-equivalent) plus EU/international tools that are under-detected by Anglo-Saxon tooling (Brevo, Crisp, Klaviyo EU, OVH, etc.).
How many sites can be analyzed in one session?
During alpha, there is no fixed cap. Allow ~0.8 to 1.5s per site (HTML up to 1.5 MB). On a list of 1,000 domains, detection typically completes in 15–25 minutes depending on average page size.
Does the tool detect "consent-before-load" tracking?
Partially. Modern SPA sites using a CMP like Didomi or OneTrust in consent-before-load mode only fire pixels after cookie acceptance. Path 2 (JS signals) catches what loads asynchronously, but some detections remain incomplete for these cases — this is a known limitation; a V2 deep mode (Playwright + cookie acceptance) is planned for a later release.
What is the difference from the ads_intelligence module?
Tech stack detection identifies which technologies are in use (CMS, analytics, e-commerce…). The ads_intelligence module goes further on the advertising/marketing axis: ad intensity, 0–100 score, categorization (retargeting, conversion, native, affiliate…), and a full list of detected trackers. The two modules are complementary.
Can results be exported as CSV?
Yes — like all outsend outputs: CSV for Excel/Sheets, JSON for programmatic use, XLSX for pre-formatted spreadsheet sharing. Detection adds named columns (cms, ecommerce, analytics, pixel_meta, etc.) to your existing list.
Is the analysis ethical and non-intrusive?
The tool reads the public HTML served by the site (just as a browser would). It makes no attempt to access authenticated areas or bypass protections. The method is equivalent to a browser visit, simply automated.