Homepage Design: What Should Visitors See First?

Before a visitor scrolls, they should be able to say three things: what you sell, whether you sell it to someone like them, and what to do next. Good homepage design is mostly the discipline of getting those three into the first screen and moving everything else down. A rotating banner, a stock photograph of a handshake and the word Welcome answer none of the three.

That first screen is worth more than the rest of the page combined, and it is usually the part a business spends least time on. Below: what belongs in it, what should move down, how the fold behaves in 2026, and a twenty-minute test for your own site.

What should a homepage show in the first screen?

Five things, in roughly this order of importance. A homepage carrying all five in the opening screen is unusual; one carrying the first three is already ahead of most.

• What the business actually does, in the words a customer would use. Not a category (“digital solutions”), not a mood (“driven by excellence”), but the thing you would say on the phone. Aircon servicing for offices in Klang Valley. Company secretarial work for Sdn Bhd start-ups.
• Who it is for, and where. Location matters more in Malaysia than most templates assume, because half the visitor’s question is whether you cover their area.
• One clear next step. One. A quote request, a WhatsApp enquiry, a booking, a call. Two buttons of equal weight amount to none, and “Learn more” is a step towards nothing in particular.
• One piece of proof. A named client, a certification, a licence number, a count of years. One real thing beats five badges nobody recognises.
• A visible way to reach a person. A tappable phone number, a WhatsApp link, an address. Buried in the footer is not visible.

The first item is where most homepages fail, and they fail in one particular way: adjectives instead of a sentence. Three one-word virtues stacked over a night-time photograph tell a visitor nothing they could not have guessed. The Nielsen Norman Group makes the same point in its homepage design principles: say who you are and what you do, show real content rather than describing it, and keep the page simple enough that the eye can finish it.

Proof is the item most often dropped, usually because it feels like bragging. Visitors run a credibility check in the first few seconds, on ordinary operational details rather than design polish — the subject of our piece on the trust signals visitors notice before they contact you.

If a stranger cannot finish the sentence “this business does ___ for ___” from your first screen, nothing further down the page will be read carefully.

Where does the fold actually sit in 2026?

The fold is not a line the browser draws. It is a change in how hard people are willing to work, and it is sharper than “people scroll now” suggests. In eyetracking research published by the Nielsen Norman Group in 2018, users spent 57% of their viewing time on the first screenful, 17% on the second, and the remaining 26% spread thinly across everything below.

Two caveats. The same team measured 80% above the fold in 2010, so the trend runs towards more scrolling, not less; and an average across many sites is not a prediction about yours. What holds is the shape: a tall first screen, a much shorter second, then a long thin tail.

So there are two folds to design for, not one, and they are different heights. On a mid-range Android phone a sticky header plus a full-bleed hero can consume the whole visible area, leaving the headline below the fold on the device the visitor is holding. On a laptop the same page often shows too much at once and the eye has nowhere to land.

Which matters more depends on your own traffic, and that is worth checking rather than assuming. StatCounter recorded Malaysian web traffic at 64.13% desktop and 34.88% mobile for August 2026 — the reverse of what most people expect, and the reverse of what many consumer sites see in their own analytics. Open your numbers, take the two commonest screen sizes, and design the first screen for both.

What does not belong in the first screen

Most homepages do not lack content above the fold. They spend it on the wrong things.

• A rotating carousel. Each slide reaches a fraction of visitors, the message changes while they are reading it, and the argument over which slide goes first should have produced one headline instead.
• A greeting. “Welcome to” plus your company name spends the most valuable line on the site saying nothing. “Your trusted partner in” spends it saying less.
• A hero image with no information in it. A large photograph earns its place when it shows the actual work, premises or product. A stock office is a wall with a view.
• Autoplay video. It costs load time, it eats mobile data, and anything running longer than five seconds needs a visible pause control to meet WCAG 2.1.
• A pop-up on arrival. A newsletter overlay shown before the visitor knows what the business does is a door closing on the way in.

None of these are new mistakes. They persist because they are what a page-builder template offers by default, and because deleting something from a homepage feels riskier than adding to it. It is the other way round.

What should come next, and in what order

After the first screen, the homepage answers the questions a real enquiry raises, in the order a customer raises them.

• Second screenful: what you do, itemised. Three to six services in plain language, each linking to a page that answers rather than describes. This is the screenful that gets the 17%, so it should not be a mission statement.
• Third: proof at scale. Named clients, a short case with a number in it, a certification with its issuing body. Logos alone persuade nobody who does not recognise them.
• Fourth: the objection. Cost range, lead time, what is excluded, coverage area. Answering the awkward question separates an enquiry from a browse.
• Fifth: the enquiry path. A short form, a WhatsApp button, a phone number, business hours. Repeat the first screen’s single action rather than introducing a new one.

The enquiry path leaks most quietly, because it fails without any visible error: forms asking for a company registration number before a name, WhatsApp links that open a blank chat, contact pages with nobody behind them. We wrote about that pattern in why a website with visitors still generates no enquiries, and the fix is nearly always subtraction.

How to test your own homepage design in twenty minutes

No tools, no budget, and you can do it this afternoon. Five checks:

• The five-second test. Show the first screen to someone outside the business for five seconds, take it away, and ask what the company does and who for. If they hesitate, the copy is the problem, not the design.
• The crop test. Screenshot the homepage on your phone and on a laptop without scrolling. Everything in those two rectangles is your homepage, as far as most visitors are concerned.
• The squint test. Half-close your eyes at the first screen. You should still see one obvious block of text and one obvious button. If everything blurs to the same weight, there is no hierarchy.
• The 4G test. Turn wifi off on a mid-range phone and load the homepage. Time how long until the headline is readable, not until the page finishes loading.
• The stranger test. Ask someone who has never heard of the business to find your phone number and your coverage area. Watch without helping, and count the scrolls.

Failing the five-second test and the squint test together usually means restructuring rather than restyling: new templates and new copy, not a new colour palette on the same layout. Failing only the 4G test is an image and hosting problem with a much smaller bill attached.

Frequently asked questions

Where iRevamps fits

Most of the homepages we are asked to look at do not need replacing. They need the first screen rewritten, the carousel removed, the enquiry path shortened and the order of the sections changed to match how customers actually decide. Our website revamp and design service starts by testing the existing homepage the way described above and reporting what it finds, before anyone talks about a new design. If the five-second test is the one you are least confident about, start there.