Performance myths: why score chasing fails and what works instead

Why score chasing fails

Performance scores are useful tools, but they are not the goal. Chasing a perfect score can lead you away from real improvements.

For more on performance myths and what actually works, see performance myths and quick fixes.

Common performance myths

Myth 1: "A score of 100 means the site is fast"

Lighthouse scores are useful, but they are lab tests [2]Source 2 . They cannot represent every device, network, or user journey.

• Lab vs field: Lab tests use controlled conditions. Real users have varied devices, networks, and locations.
• Single page focus: Scores measure individual pages, not complete user journeys.
• Perfect conditions: Lab tests use fast connections and powerful devices. Many users do not.

A perfect score in the lab does not guarantee fast performance for real users.

Myth 2: "A plugin will fix performance"

Performance plugins promise quick fixes, but the biggest wins nearly always come from fundamental changes.

• Reducing payload: Smaller pages load faster. Remove unnecessary code, images, and assets.
• Cutting third-party scripts: Third-party scripts add weight and delay [5]Source 5 . Remove what you do not need.
• Fixing render blocking assets: CSS and JavaScript that block rendering delay when content appears.

Plugins can help, but they cannot fix fundamental problems with page weight and loading order.

Myth 3: "Core Web Vitals are only for SEO"

Core Web Vitals are user experience signals [1]Source 1 . Google defines and documents them as real-world performance metrics.

• LCP (Largest Contentful Paint): How quickly the main content appears. Users see this directly.
• INP (Interaction to Next Paint): How quickly interactions respond. Users feel this directly.
• CLS (Cumulative Layout Shift): How stable the page is while loading. Users notice this directly.

Improving Core Web Vitals improves real user experience, not just search rankings.

Myth 4: "Quick fixes are enough"

Quick fixes can help, but sustainable performance comes from good foundations.

• Image optimisation: Important, but only one part of performance.
• Caching: Helps, but does not fix slow initial loads.
• CDN: Can help, but does not reduce page weight.

Quick fixes are useful, but they work best when combined with fundamental improvements.

What works instead

1) Measure lab and field

Use both lab tests and real user data to understand performance [3]Source 3 .

• Lab tests: Lighthouse for controlled testing and identifying issues.
• Field data: Real user metrics from CrUX (Chrome User Experience Report) where available.
• Combine both: Lab tests find problems. Field data shows real impact.

2) Set budgets

Set clear limits for page weight, request count, and third-party scripts.

• Page weight: Target maximum total page size (e.g., under 1MB for mobile).
• Request count: Limit the number of HTTP requests per page.
• Third-party limits: Restrict third-party scripts to what is essential.

Budgets help prevent performance from drifting over time.

3) Fix the big offenders first

Prioritise fixes that make the biggest difference to real users.

• Images: Optimise images, use modern formats, lazy load where appropriate.
• Fonts: Load fonts efficiently, avoid invisible text during load.
• Scripts: Remove unnecessary scripts, defer non-critical JavaScript.
• Render blocking CSS: Inline critical CSS, defer non-critical styles.

Focus on what users experience, not just what scores measure.

4) Re-test and document changes

Test before and after changes, using the same test conditions.

• Before metrics: Document current performance with specific numbers.
• After metrics: Measure again after changes to see improvement.
• Same conditions: Use the same device, network, and test setup for fair comparison.

Documentation helps you understand what worked and what did not.

Real user experience matters most

Performance is about how your site feels to real users [4]Source 4 , not about achieving perfect scores.

• Mobile first: Most users are on mobile. Test and optimise for mobile connections.
• Real journeys: Optimise complete user journeys, not just individual pages.
• User perception: Fast feels fast. Focus on what users experience, not just metrics.

Summary

Performance myths: perfect scores mean fast sites, plugins fix everything, Core Web Vitals are only for SEO, quick fixes are enough.

What works: measure lab and field, set budgets, fix big offenders first, re-test and document changes.

Real user experience matters most. Focus on how your site feels to real users, not just scores.

For more on performance myths and what actually works, see performance myths and quick fixes. For help improving performance, see performance services or fix a slow website. For more on what fast means, see fast websites: what fast means in 2026. You can also get in touch to discuss your performance needs.