Skip to content
Science-backed data on why AI says what it says
Browser usability

Layout shift

Layout shift is movement of visible page content between rendered frames. Cumulative Layout Shift, or CLS, summarises unexpected movement. BLURSOR shows a bounded browser observation rather than a field measurement of the whole visitor experience.

Checked against BLURSOR’s measurements ·

What it means

A late-loading banner or image can move the paragraph someone is reading or the button they intend to press. The useful question is which element moved, when it moved, and whether the movement interrupted the journey.

The standard CLS metric uses the largest qualifying session window of shifts. Its commonly used boundaries are 0.1 or less for good and above 0.25 for poor. Field assessment uses a population of visits, normally the 75th percentile, rather than one automated page load.

Illustrative example

Hypothetical example: a late product image

Imagine a product page that initially displays its price and enquiry button near the top. When a large image loads above them without reserved space, both move down. Reserving the image's expected space lets the content begin in its settled position.

This illustrates a cause to investigate, not a claim that a particular image caused a recorded score. A recording of the real load can show whether the movement came from the image, a banner, a font change, or something else.

How BLURSOR measures it

In the report: Page movement while loading

BLURSOR's browser probe reads layout-shift entries, excludes entries marked as following recent input, and sums the values it observes during the capture. It does not implement the standard session-window aggregation or collect field visits. Interpret the displayed CLS label as the checker's short observation.

A supported probe with no recorded shifts can return zero. Missing measurement support or missing usable browser evidence remains unknown. If CLS cannot be measured, a separate markup check can flag images without detected dimensions; that hint is not a substitute CLS value.

What to do next

  1. Record a representative page load in browser performance tools and identify the elements that moved before changing the template.
  2. Reserve space for images, embeds, and predictable late content; review font loading and banners that insert content above the visitor's position.
  3. Compare controlled tests with field data when available, and check the mobile layout and longer journeys where later shifts may appear.

What it cannot tell you

  • A zero in this capture does not demonstrate that every visit is stable. Timing, device, consent state, and later interactions can change the experience.
  • The checker result is not a Core Web Vitals certification, an accessibility audit, or proof that a person or automated agent clicked successfully.

Sources and further reading

The definitions below describe the underlying concepts. The measurement section describes BLURSOR’s own checks.

Related explanations