Documentation

Scoring & methodology

What the score means, which areas feed into it and how to read results correctly.

What goes into the score

The assessment is based on several layers:

  • server response, for example via TTFB
  • visible page build-up, for example via FCP and LCP
  • browser runtime, for example via TBT
  • layout stability, for example via CLS
  • requests and payload size
  • caching, images, assets and technical delivery
  • further WordPress and front-end signals that become visible in the report

The important part: it is not one number that counts, but a combination of metrics, runtime and detected bottlenecks.

Score bands

In the product, score values are divided into four bands:

Value Band Meaning
90–100 Very good Technically very solid. Individual bottlenecks are still worth keeping an eye on.
75–89 Good Well set up overall. Check individual bottlenecks and recommendations.
60–74 Room for improvement Clearly visible bottlenecks in load time, rendering or delivery.
0–59 Critical Several problems are slowing the page down noticeably. Prioritised measures make sense.

You will find these bands in scan and report views as well as in alerts and domain overviews.

Building blocks of the sub-scores

The overall score is made up of several sub-areas:

Speed

Measures server response, visible page build-up and browser runtime. The important signals are TTFB, FCP, LCP and TBT.

Images

Assesses image optimisation, formats (WebP, AVIF), missing dimensions and lazy loading.

Caching

Checks HTTP cache headers, whether static resources are cached sensibly and whether the server responds quickly.

Frontend

Blocking scripts, DOM size and the resource breakdown show how much JavaScript and CSS slow down the page build-up and how much unnecessary load is transferred.

Fonts

Fonts directly affect how visible text builds up. Outdated formats such as TTF or EOT, missing WOFF2 variants, a missing font-display and large font files all feed into this sub-score.

WordPress

WordPress-specific checks: active plugins, theme quality, known bottlenecks from the WordPress ecosystem.

How to read the score

A single score value is a starting point, not a verdict. Only in combination with the sub-areas, bottlenecks and problems does it become clear why a page has that value.

Why the score alone is not enough

A similar overall score can have very different causes. That is why these parts of the report are often particularly important:

  • the score breakdown
  • the biggest bottlenecks
  • problems
  • recommendations
  • baseline and comparison

They help you understand the direction of a problem instead of just judging the number.

Why values can fluctuate

Results can change because of:

  • region and chosen location
  • differences between mobile and desktop
  • time of day and load
  • cache state
  • third-party scripts
  • CDN, WAF or bot protection
  • dynamic content

That is why a single scan is always just a snapshot.

Why repeated scans matter so much

Only repeated or scheduled scans show whether a page stays reliably fast or just happened to catch a good or a bad moment.

That is exactly why history, baseline and alerts are often more important for real work than a single scan without context.

What the score is and what it is not

What it is

A technical assessment that makes bottlenecks visible, helps prioritise measures and makes changes over time understandable.

What it is not

Not an SEO ranking, not a guarantee and no substitute for looking at problems, recommendations and history.

Google PageSpeed score

Every scan additionally shows a Google PageSpeed score for mobile and desktop. This value comes from the Google PageSpeed Insights API and is based on Google Lighthouse — it is fetched independently of the turbometrics score and is visible in the scan report on the details tab.

Value Band (Google PSI)
≥ 90 Very performant
50–89 Room for optimisation
< 50 Slow — action needed

The Google PageSpeed score complements the turbometrics score. Both measure synthetically under lab conditions, but they weight things differently. Differences between the two values are normal and not a contradiction — they shed light on similar aspects from different perspectives.

Further reading

MetricsProblems & recommendationsHow we score