Zero measurable impact on your store

Detection you can't feel. The browser snippet stays off the critical render path, and server-side detection does nothing at all for human visitors. Here's exactly why, and how to reproduce it yourself.

01The browser snippet

• Under 5 KB gzipped, loaded with async + defer — never blocks parsing or rendering.
• Reports via sendBeacon — fires off the critical path, doesn't hold the main thread, survives page unload without blocking navigation.
• Zero DOM manipulation. It observes; it never writes to the page, so there's no layout shift and no reflow.
• Classification happens at Tactical's edge, not in the browser — no scoring work runs on your visitor's device.

02Server-side detection (WooCommerce plugin / middleware)

• Matched-only firing. An event is created only when the request's user-agent matches a known AI crawler. Humans and unknown UAs cost a single in-memory lookup and nothing else — no HTTP call, no database write.
• Non-blocking delivery. The report is a fire-and-forget request with a hard 1 second timeout; the visitor's response never waits on it.
• Burst-guarded. A short per-(agent, URL) dedup absorbs crawler storms before they leave the site.

Net effect: a human page load incurs no extra outbound request and no measurable TTFB change. The only requests Tactical makes are for the AI traffic you're trying to measure.

03Reproducible Lighthouse test

Don't take our word for it — run it:

1. Open the demo store in an incognito window.
2. Chrome DevTools → Lighthouse → Mobile → Performance. Run 3× with the Tactical snippet disabled; take the median.
3. Enable the snippet. Run 3× again; take the median.
4. Compare Performance score, LCP, TBT, and CLS.

Published scores from our demo store go here — the delta should be within run-to-run noise. We publish the methodology so you can reproduce it, not just the numbers.