Total Blocking Time (TBT) Metric
What you'll learn: You'll understand how TBT measures the cumulative effect of long tasks that delay user interactions.
What is Total Blocking Time?
Total Blocking Time (TBT) is a Core Web Vital that quantifies how much your page's main thread is blocked during the critical loading period. It specifically measures the sum of all blocking time from long tasks between First Contentful Paint (FCP) and Time to Interactive (TTI).
Here's the key: any task that runs longer than 50ms is considered a "long task," and TBT adds up only the excess time beyond that 50ms threshold.
How TBT is Calculated
Imagine three long tasks during page load:
- Task 1: 70ms (20ms blocking time)
- Task 2: 120ms (70ms blocking time)
- Task 3: 60ms (10ms blocking time)
TBT = 20ms + 70ms + 10ms = 100ms
The lower your TBT, the more responsive your page feels. A TBT under 200ms is considered "good," while anything over 600ms needs improvement.
Why TBT Matters for Interactivity
Think of TBT as measuring how long users are "locked out" of interacting with your page. When a long task runs, clicks, taps, and keyboard inputs are queued but not processed. TBT captures this frustration numerically—it's the cumulative "wait time" users experience when trying to interact during critical moments.
Unlike just counting long tasks, TBT weights them by severity: a 200ms task contributes more to user frustration than a 60ms task.
Connection to Previous Concepts
You've already learned how to identify long tasks with the Long Tasks API and Performance API, and techniques like setTimeout and scheduler.yield() to break them up. TBT is the metric that proves whether those optimizations actually improved user experience.
Key Takeaway: Total Blocking Time measures the cumulative impact of long tasks on interactivity by summing the blocking time (beyond 50ms) of all long tasks during initial load—lower TBT means faster, more responsive interactions.