9 Website Performance Metrics

Website performance is easier to fix when each metric is tied to a real bottleneck. Some numbers explain server delay, some explain transfer cost, and some explain why the browser still feels slow after bytes arrive.

1. Load time

Load time is the total time until the browser considers the page fully loaded. It is a broad regression signal, but not a complete user-experience metric because a page can finish loading late or early relative to when useful content actually appears.

2. Time to First Byte

Time to First Byte, or TTFB, measures the delay until the first response byte reaches the browser. It captures DNS, connection setup, TLS, edge routing, and backend work, which makes it a strong signal for server-side latency or cache misses.

3. Request count

Request count is the number of network requests needed to assemble the page. Too many requests increase overhead even on modern protocols, and they often point to fragmented bundles, excessive third-party tags, or too many small assets.

4. DOMContentLoaded

DOMContentLoaded fires when the initial HTML has been parsed and the DOM is ready. It is useful for spotting heavy synchronous scripts or oversized documents, but it is only a parsing milestone, not proof that the page is visually ready.

5. Above-the-fold load time

Above-the-fold load time measures how quickly the content in the initial viewport appears. It matters because users judge the page before they scroll, so slow hero content often hurts perceived speed more than a slower footer or later image.

6. First Contentful Paint

First Contentful Paint, or FCP, records when the browser first paints visible DOM content such as text or an image. It is a clear signal that something reached the screen, but it can still look good while meaningful content remains delayed.

7. Page size

Page size is the total byte weight of the page and its dependent assets. Large pages take longer to transfer and often longer to parse and render, so image formats, compression, bundle size, fonts, and cacheability all show up here.

8. Round Trip Time

Round Trip Time, or RTT, is the time for a request to travel to the remote endpoint and for a response to return. It explains why a page can feel slow even with an efficient backend, especially when users are far from the origin or the application makes too many sequential requests.

9. Render-blocking resources

Render-blocking resources are files, usually CSS and synchronous JavaScript, that prevent the browser from painting useful content. A few badly placed assets can delay the whole critical path, which is why critical CSS, script deferral, and third-party tag control matter.

Read them as a set

TTFB tells you where the wait starts. Request count, page size, and RTT explain transfer cost. DOMContentLoaded, FCP, above-the-fold timing, and render-blocking resources explain what the browser is doing with the bytes it received. Used together, these metrics turn vague slowness into a diagnosable system.