What Is Above the Fold?
Above the fold is the part of a web page a visitor sees on arrival, before any scrolling happens. It is not a fixed measurement, and the boundary moves.
Above the fold is the part of a web page a visitor sees on arrival, before any scrolling happens. It is not a fixed measurement. The boundary moves with screen size, browser chrome, zoom level, font settings, and whatever consent banner loads on top, so one page has as many folds as it has visitors.
Why the first screen still decides things
Attention is spent before it is earned. A visitor arriving from search is checking one thing: whether this page answers the question they just typed. If the first screen shows a stock photograph, a rotating banner, and a headline that could belong to any competitor, they have no reason to scroll, and no amount of excellent writing further down will be read.
The damage compounds quietly. Rankings can hold while the page converts nothing, because the search engine sees a relevant document and the visitor sees decoration. Teams then chase links and word count, when the actual fault is that the first 400 pixels never confirm the visitor is in the right place. If your top-of-page copy is doing no work, it is worth fixing what people read first before you touch anything below it.
What actually determines where the fold falls
No browser sends you a fold position. What exists is a viewport, and everything that competes for space inside it. The term comes from newspapers, where a broadsheet was folded in half on the stand and only the top half faced the buyer. That constraint was physical and identical for every reader. On the web, nothing about it is fixed, and the variables below are what decide what a given person sees first.
- Viewport height, not screen height. The usable area is the screen minus the browser interface, and on mobile that interface expands and contracts as the user scrolls.
- Zoom and font preferences. A visitor using larger text sees noticeably less content in the same space, and accessibility settings are more common than most teams assume.
- Consent and cookie banners. A banner that covers the lower half of a phone screen has effectively raised the fold, and it appears before your content finishes rendering.
- Sticky headers and floating elements. Every pixel of persistent navigation, promotional bar, or chat widget is a pixel your opening paragraph does not get.
- Render order. What is visible at 400 milliseconds may differ from the final layout, because images, fonts, and injected scripts arrive at different moments.
- Layout shift after load. If elements move once late assets land, the content someone started reading can be pushed out of view entirely.
This is why arguments about the fold rarely resolve. Two people describing the same page are describing two different screens, and both are correct. The measurable consequence shows up elsewhere, in the share of visitors who arrive and leave without engaging, which is one of the things a bounce rate figure is measuring when the number is defined properly.
Replacing the argument with a measurement
| The unanswerable question | The version you can actually check |
|---|---|
| Is the call to action above the fold? | At which viewport heights does it become visible without scrolling? |
| Is the hero image too big? | Is that image the largest contentful paint element, and how long does it take? |
| Do people see our value proposition? | What text is painted in the first viewport on a mid-range phone? |
| Does the banner hurt us? | How much viewport height does the consent layer occupy on a small screen? |
| Is the page too slow at the top? | What is the field largest contentful paint against Google’s 2.5 second threshold? |
The blunt version
There is no single fold, so stop designing to one. The useful question is narrower and it has an answer: what renders in your largest contentful paint. Largest contentful paint (LCP) is the Core Web Vitals metric that records when the biggest text block or image in the initial viewport finishes rendering, and Google documents 2.5 seconds as the threshold for a good score.
That metric tells you two things a fold debate never can. It names the element the browser considers the main visual payload, and it times it. If your LCP element is a decorative hero image, your page is spending its opening moment on something no visitor came for. If it is a headline that states the answer, the opening is working.
The industry keeps the vaguer version alive because vagueness is easier to sell. A designer can defend any layout by choosing a favourable screen size, and an agency can promise an above-the-fold optimisation without ever stating what changed. Ask for the LCP element instead. Ask what it is, how long it takes, and whether it answers the query the visitor typed. That question forces a specific answer, which is also the standard applied to the rest of the page in a proper on-page SEO checklist.
Example
Say a recruitment agency runs a landing page for temporary warehouse staff. On a designer’s large monitor, the headline, the sub-headline, and the enquiry form all sit comfortably in view. On a mid-range phone, the first screen holds a sticky header, a full-width photograph of a forklift, and the top third of the headline, with a consent banner covering the rest. Field data shows the forklift image as the largest contentful paint element, arriving slowly. Nobody on the team saw a problem, because nobody tested at the size most visitors use. The fix costs nothing structural: shrink the image, promote the headline and the availability line above it, and let the picture load after the words that do the persuading.
FAQ
Does Google rank pages by what is above the fold?
Not as a named factor. Google measures largest contentful paint as part of Core Web Vitals, and it has a documented policy on intrusive interstitials that cover content. Those are specific, published behaviours. An abstract fold position is not something the ranking systems are documented to measure at all.
Where is the fold on a mobile phone?
There is no fixed position, because the browser interface resizes as the user scrolls and device sizes vary widely. Test a range of viewport heights rather than choosing one. The practical approach is to make the first screen work at the smallest common size you see in your own analytics.
Should the call to action always be above the fold?
No. For a low-commitment action such as a search or a newsletter box, early placement helps. For anything requiring confidence, such as a quote request, asking before you have earned attention performs worse. Place the action where the argument for it is complete, not where the screen happens to end.
Related terms
- Share of Voice — the visibility metric that says nothing about whether arriving visitors stay.
- Log File Analysis — the server-side record of crawler behaviour, which no rendering test can replace.
- Bounce rate — the number people quote when a first screen is failing, usually without defining it.
Nobody can tell you where the fold is, but anyone can tell you what your largest contentful paint element is. Ask for the second thing, and the first argument disappears.