JavaScript execution time
Silktide measures how much time the browser spends running scripts while your page loads. Heavy script work keeps the page from responding, so a page can look loaded yet ignore taps and clicks for seconds.
Why this matters
Downloading a script is only the first cost. The browser must then parse and run it, and all of that happens on the same thread that handles scrolling, typing, and taps. While scripts run, the page is deaf: buttons do not respond and the page feels broken even though it looks finished.
Unlike download time, this cost scales with the visitor's processor - a script that runs in half a second on a developer's machine can take several seconds on a budget phone. The visitors with the cheapest devices wait longest.
How to fix it
- Ship less . Removing unused code is the most direct win - see Unused JavaScript.
- Split large bundles and load code for below-the-fold or interaction-driven features only when needed, so the initial load runs only what the first screen requires.
- Audit third-party scripts, which often dominate execution time - see Third-party code. Remove those whose value does not justify their cost.
- Move genuinely heavy computation into a web worker, which runs on a separate thread and leaves the page responsive.
How Silktide tests this
- Load the page in a real browser, using the device type, connection speed, and test location configured for your speed tests.
- Measure the processing time each script consumed during the load, split into running the code (script evaluation) and preparing it to run (parsing).
- Report the scripts that consumed meaningful time, with a per-file breakdown. When total execution time is low, no individual scripts are listed.
- Grade the page by the total: under 2 seconds is good, and around 3.5 seconds or more is graded as poor.
Troubleshooting
A third-party script tops the list
Analytics, tag managers, ads, and chat widgets frequently cost more execution time than a site's own code. You cannot optimise their internals, but you can load them later, replace them with lighter alternatives, or drop the ones nobody uses.
The page feels fast on my machine
Execution time depends on the processor. Developer machines are typically several times faster than the mid-range phones most visitors use, so a budget that feels invisible to you can be seconds for them. The measured figures come from a controlled test, not your hardware.