Web Performance9 min readAugust 18, 2026

How to Inspect & Debug Web Performance Bottlenecks with HAR Files

A comprehensive guide on recording, inspecting, and analyzing HTTP Archive (HAR) files. Learn how to pinpoint slow TTFB, identify render-blocking assets, analyze waterfall charts, and sanitize sensitive cookies.

ToolBean Performance Lab

ToolBean Performance Lab

Frontend Performance Specialists

Try the Companion Tool

Online HAR File Viewer

Analyze HTTP Archive logs, inspect network waterfall charts, and find slow API requests with zero data upload.

Launch Tool

1. What is a HAR (HTTP Archive) File?

A HAR (HTTP Archive) file is a standardized JSON-formatted file used by all major web browsers to capture and track detailed performance metrics of every network request made during a page session. It includes DNS resolution time, SSL handshake duration, request/response headers, status codes, cookies, and transfer payloads.

2. Understanding the Request Lifecycle Breakdown

When you inspect a request in our HAR Viewer, the total duration is divided into distinct phases:

  • Blocked / Queued: The time the request spent waiting in the browser queue (e.g., waiting for an available TCP socket or higher-priority assets).
  • DNS Lookup: Time taken to resolve the domain name to an IP address. Can be optimized using <link rel="dns-prefetch">.
  • Connect / SSL: Time spent completing the TCP three-way handshake and TLS/SSL certificate negotiation.
  • Send: Time required to send HTTP request headers and body to the server.
  • Wait (TTFB): Time spent waiting for the server to process the request and return the initial byte.
  • Receive (Content Download): Time spent downloading the response body. High download time points to uncompressed images, large JavaScript bundles, or bandwidth bottlenecks.

3. Common Performance Bottlenecks Revealed by HAR Analysis

  1. Sequential Waterfall Chains: Look for cascading requests where request B doesn't start until request A finishes. This often happens with dynamic @import in CSS or chained API calls.
  2. Uncompressed Assets: Check the Content-Encoding header. If text resources (HTML, JS, CSS, JSON) lack gzip or br (Brotli) compression, payload sizes can be 70% larger than necessary.
  3. Missing Cache-Control Headers: Static assets (logos, fonts, scripts) with Cache-Control: no-cache force repeated downloads on every page reload.
  4. Unoptimized HTTP Status Codes: Frequent 301/302 redirects add an extra full network roundtrip before the real resource can even begin loading.

4. Privacy Best Practice: Sanitizing HAR Logs

Because HAR files capture authentication cookies, Bearer tokens, and user credentials, using an in-browser, client-side HAR viewer like ToolBean ensures your proprietary logs remain on your device and are never sent to external servers.

Frequently Asked Questions

HAR files record the entire raw HTTP conversation, which includes session cookies, Authorization bearer tokens, API keys in headers, and POST body data. Always sanitize HAR files before sharing them publicly.

Recommended Guides