Lighthouse Simulator

Use our Lighthouse Simulator to forecast web performance. Adjust LCP, TBT, and CLS to optimize Lighthouse metrics and improve your site's Core Web Vitals score.

xDevToolsInitializing Tool

Related Utilities

Last Updated: August 14, 2026|Author: Yogeesh S, Senior Software Engineer

How the Lighthouse Metrics Scoring Algorithm Works

The performance scoring engine in this simulator mirrors the standardized log-normal distribution used by industry-standard audits to calculate your final grade. Each of the five core pillars—First Contentful Paint (FCP), Speed Index (SI), Largest Contentful Paint (LCP), Total Blocking Time (TBT), and Cumulative Layout Shift (CLS)—is weighted according to its impact on user experience.

TBT carries the highest weight at 30%, followed by LCP and CLS at 25% each, while FCP and SI contribute 10% apiece. The math relies on specific median and "poor" threshold values to normalize these distinct units—seconds and milliseconds—into a single 0-100 scale. By understanding these weights, you can prioritize which optimizations will yield the highest return on investment for your overall web performance.

Interpreting Your Lighthouse Metrics Diagnostics

When your simulation results fall below the ideal range, the diagnostic panel provides specific, actionable feedback based on the values you have set. If you see a "Needs Improvement" or "Poor" status, it indicates that your input parameters have crossed the thresholds that Google's audit engine uses to flag performance bottlenecks.

These diagnostics are not just alerts; they are direct technical recommendations for your critical rendering path. For instance, if your LCP is flagged, the tool suggests focusing on asset delivery—specifically, how your largest above-the-fold elements are prioritized by the browser. Similarly, a high TBT warning points to the need for code-splitting or offloading main-thread execution, which is critical for maintaining a responsive interface.

Customizing Your Web Performance Simulation Controls

The simulation workspace features five primary sliders, each representing a critical metric, allowing you to model various scenarios without running a live audit. Each control is mapped to a specific range that reflects real-world network and device conditions.

MetricRangeImpact
LCP0.5s – 8.0sMeasures perceived load speed of the main content
TBT0ms – 1200msTracks total duration of main thread blocks during load
CLS0.0 – 0.6Quantifies visual stability and layout shifts
FCP0.5s – 5.0sIndicates when the first piece of content renders
SI0.5s – 10.0sCalculates how quickly content is visually populated

Step-by-Step Guide to Modeling Core Web Vitals

1

Adjust the LCP Slider

Drag the control to match your current Largest Contentful Paint. If your banner image takes 3.0 seconds to render, set it there to see the immediate impact on your projected score.

2

Refine the TBT and CLS Inputs

Use the sliders to reflect your current main-thread blocking time and layout shift indices. These values are often the "silent killers" of a high score.

3

Review the Performance Gauge

Watch the circular score indicator update in real-time as you drag the sliders. A score of 90 or above indicates a "Good" rating.

4

Action the Diagnostic Tips

Read the generated text below the gauge to find specific fixes for your current configuration, such as implementing priority hints or reserving layout space for images.

Practical Example: Improving a Sub-Optimal Page

Imagine you have a page with an LCP of 3.2 seconds and a TBT of 450ms. Using the simulator, you can see that this combination likely drops your score into the "Improve" category.

By adjusting the LCP slider down to 2.2 seconds (via preloading) and reducing TBT to 150ms (by deferring non-critical JavaScript), you can instantly visualize how these changes push your score back into the "Good" (90+) range. This allows you to plan your development sprint based on measurable projected outcomes rather than guesswork.

Best Settings for Realistic Performance Forecasting

When using the simulator to forecast future state, always start by inputting your current real-user monitoring (RUM) data rather than arbitrary numbers. For mobile-first optimization, prioritize setting your TBT higher, as lower-end mobile CPUs struggle substantially more with main-thread execution than desktop environments.

If you are aiming for a perfect 100, focus on keeping your LCP under 2.5 seconds and your TBT under 200ms, as these two metrics are the most frequent culprits for lower scores. Use the "Reset" button to return to baseline settings if you lose track of your modifications during your experimentation phase.

FAQ: Resolving Discrepancies in Lighthouse Metrics Simulations

Why does my simulation score differ from my actual live site audit?

The simulator uses a simplified log-normal approximation for speed, whereas a live audit accounts for network jitter, device CPU throttling, and server response time variations. Treat the result here as a projection for optimization planning rather than an exact replica of a field report.

When should I prioritize LCP over TBT for performance gains?

Prioritize LCP if your users are bouncing before the main content even appears. Prioritize TBT if your page loads visually but remains unresponsive to clicks or scrolls, which usually suggests excessive JavaScript execution.

How can I simulate different device profiles in this tool?

This simulator focuses on the metric values themselves. To model specific device profiles, simply adjust the input sliders to represent lower-end hardware, where FCP and TBT are naturally higher.

Which Core Web Vitals metric causes the most consistent score drops?

Total Blocking Time (TBT) is the most frequent cause for score degradation in current applications. Because it carries a 30% weight, even small reductions in blocking code can produce significant score increases.

Does this simulator account for the latest algorithm updates?

Yes, the scoring weights are based on current standards, including the 30% TBT weight and 25% LCP and CLS weight distribution.

Can I use these projections to justify performance budget requirements to my team?

Absolutely. By demonstrating that a 1-second reduction in LCP can move your score from 60 to 85, you provide quantitative evidence for the necessity of performance-focused development tasks.

Why is my CLS score low even though nothing seems to move on the page?

CLS can be impacted by dynamic ad injection or late-loading fonts that cause reflows. Even if you don't see it, ensure all containers have fixed dimensions to prevent invisible shifts.

Is there a way to save my specific simulation configuration?

Currently, the tool resets on page refresh. To keep track of your performance goals, note down the slider values that yield your target score and share those with your developers as a performance budget.