Necessary Elements of a Great Website

The most useful web design tip is unglamorous: decide what the site has to accomplish before deciding how it should look. Almost every failing business website is failing because that question was never answered, not because the design was unfashionable.

What follows is the set of decisions that determine whether a business website works, roughly in the order they should be made.

Settle the job before the design

A business website usually has one primary job and two or three secondary ones. Enquiries, applications, bookings, sales, or simply being credible enough that people who found you elsewhere feel comfortable proceeding.

Write it down. "Prospects who were referred to us should be able to confirm we are legitimate and get in touch within two minutes" is a job you can design against. "A modern site that reflects our brand" is not.

This single sentence resolves most later arguments, because it gives you a criterion. Every proposed element can be tested against it.

Say what you do, immediately

The most common failure on business websites is that a visitor cannot tell what the company does within a few seconds of arriving.

Your homepage needs a plain description of what you offer and who it is for, visible without scrolling, in language a customer would use. Not a slogan, not a mission statement, and not industry vocabulary that only makes sense internally.

A useful test: show your homepage to someone outside your industry for five seconds, then ask what the company does. If they cannot say, no amount of visual quality compensates.

Structure before styling

How content is organised determines whether people find things, and it is decided before any visual work.

Navigation that matches how people think

Label navigation with the words your customers use, not your internal department names. If people search for "repairs" and your menu says "aftercare solutions", you have added a translation step.

Keep the top level short — around five to seven items. Menus with fifteen items are a filing cabinet, and everything in them is equally invisible.

Keep important pages shallow

Anything commercially important should be reachable in two clicks from the homepage. Pages buried three levels deep receive little traffic and little internal link value, however good they are.

One page per thing you sell

A single page listing eight services ranks for none of them and answers none of them properly. Each distinct offering deserves its own page with its own detail.

Design for scanning

People do not read business websites. They scan until something matches what they need, then read that.

• Clear visual hierarchy. Deciding what the eye reaches first, second and third is the working part of visual design. Emphasising everything emphasises nothing.
• Descriptive headings. Someone reading only your headings should understand the page.
• Short paragraphs. Three or four lines. Dense blocks get skipped.
• Constrained line length. Around 60 to 75 characters. Full-width text on a wide monitor is genuinely tiring to read.
• Real whitespace. Space around elements is what makes them legible. Filling every gap makes a page harder to use, not more informative.

Make the next step obvious

Every page should have one clear primary action, and it should be the one that page is for. A service page invites an enquiry; a case study invites a conversation about similar work.

Label buttons with what happens — "Get a quote", "Book a call" — rather than "Submit" or "Click here". And reserve one colour for the primary action so it is visually distinct from everything else, which is one of the reasons choosing colours is a functional decision rather than a stylistic one.

Trust signals, placed where doubt occurs

Business websites are checked by people deciding whether to risk time or money on you. What reassures them is specificity.

• Real photographs of real people. Stock imagery of a generic team signals the opposite of what it intends.
• Named testimonials from identifiable people at identifiable organisations. Anonymous praise carries almost no weight.
• Concrete evidence — actual clients, actual work, actual numbers you can substantiate.
• A physical address and a phone number. Their absence raises questions.
• Relevant credentials, where they mean something in your field.

Place these beside the claims they support rather than gathering them into a testimonials page nobody visits. A concern about reliability is answered by proof about reliability, at the point the concern arises.

One rule worth holding to: never publish a result, review or figure you could not substantiate if challenged. Beyond the ethics, unverifiable specifics do not persuade sceptical readers and become expensive to remove later.

Make contact effortless

A surprising amount of business is lost to friction rather than persuasion.

Contact details should be visible on every page, not only on a contact page. Offer more than one route — some people will phone, others will never phone. Keep forms short, asking only for what you will act on. Set expectations below the button: what happens next and how quickly.

And make sure someone monitors whatever you offer. A form nobody checks or a chat widget nobody staffs is worse than not offering it, because it converts an interested prospect into an ignored one.

It has to work on a phone

Most first visits to a business website are on a phone, and most business websites are designed on a large monitor.

Check the essentials on a real device: text readable without zooming, tap targets large enough and adequately spaced, no horizontal scrolling, forms that trigger the right keyboard, and the primary action reachable without hunting. The full method is in making it work on a phone.

Speed and accessibility are design decisions

Both are determined largely by choices made during design rather than by development afterwards.

A design specifying four web fonts, full-bleed video and a carousel on every page has set a performance budget regardless of how well it is built. Similarly, contrast, target sizes, focus visibility and whether meaning is carried by colour alone are all decided in design — and retrofitting accessibility is far more expensive than designing it in.

Neither is a specialist add-on. They are consequences of ordinary design choices, which is why they belong in the brief. Understanding design and development responsibilities helps here, because these two are precisely where the handoff usually drops them.

Both have published standards to design against rather than opinions to argue about. The W3C's accessibility guidelines set out the contrast, target size and keyboard requirements, and Core Web Vitals give measurable performance thresholds. Naming both in a brief converts two things that are usually hoped for into two things that can be checked.

The pages businesses consistently underinvest in

Effort goes into the homepage. The pages that actually convert usually get less attention than they deserve.

Service pages

These are where buyers make decisions and where most commercial search traffic lands, yet they are frequently three paragraphs and a stock photo. A service page should explain what the service includes, who it suits, how it works, what happens after enquiry, and answer the objections that come up in every sales conversation.

The about page

More visited than most teams expect, particularly by people deciding whether to trust you. It should say who is actually behind the business — real names, real photographs, genuine background. Corporate history nobody asked for is not the same as establishing credibility.

The contact page

Often the last page before an enquiry and frequently the most neglected. It should offer more than one route, set expectations about response time, and confirm what happens after someone submits. A bare form is a missed opportunity at the point of highest intent.

The 404 page

Someone has failed to find what they wanted. A default error message loses them. A useful 404 acknowledges the problem, offers search, and links to the main sections. It costs an hour and recovers visitors who would otherwise leave.

Design so the site can be maintained

Business websites are rarely rebuilt because they broke. They are rebuilt because they became impossible to update without a developer, so they stopped being updated and drifted out of date.

Two decisions prevent this. Ask for a component system rather than a set of unique page layouts — reusable blocks the team can assemble into new pages. And make sure someone internally is trained to add a page, update copy and change an image, and that they have done each once before handover ends.

A site the team is confident editing stays current. A site they are afraid to touch is one that needs replacing in three years for reasons that have nothing to do with its design.

What to avoid

Homepage carousels. They let a team avoid deciding what matters most. People rarely see slide two.

Autoplaying video with sound.

Interstitials on arrival. A modal asking for an email before delivering anything asks for commitment before value.

Text over unmodified photographs. Contrast varies across the image, so it fails somewhere.

Removing focus indicators. Makes the site unusable by keyboard for an aesthetic preference.

Hiding content on mobile. Content hidden from phones is hidden from search engines, which index the mobile version.

Designing with placeholder text. Real headlines are longer and less convenient than the ones in the mockup.

Sticky elements that eat the screen. A fixed header, a chat bubble, a cookie banner and a promotional bar can consume half a phone viewport before any content appears. Each was added reasonably; together they are the problem.

Getting useful feedback on a design

The review stage is where good designs get worse. Feedback gathered from several stakeholders with no framework produces contradictory instructions, and the result is a compromise nobody chose.

Review against the job, not against taste

Return to the sentence you wrote at the start. The question for every review is whether the design serves that job — not whether each person likes it. This converts a discussion about preference into one about criteria.

Describe problems, not solutions

"Make the logo bigger" is a solution to an unstated problem. "It took me a moment to work out whose site this is" is the problem, and a designer can address it in ways you had not thought of. Feedback framed as a problem consistently produces better outcomes than feedback framed as an instruction.

One person decides

Collect input from several people if useful, but one person must own the decision. Design by consensus produces the least objectionable option rather than the most effective one, and it slows everything down.

Judge it with real content

A layout reviewed with placeholder text and ideal-length headlines will look different once your actual eleven-word service name is in it. Ask to see key templates populated with genuine content before signing off — it is the single most useful thing you can ask for at review stage.

Website design requirements: a briefing checklist

If you are commissioning work, settle these before design begins. Every one that is left open becomes a decision someone else makes for you, later, under time pressure.

Purpose

• The primary job of the site, in one sentence
• The primary audience, described by situation rather than demographics
• The single action you most want visitors to take
• How success will be measured, and when it will be checked

Content

• Page inventory — what exists, what is new, what is being retired
• Who writes the copy, and by when
• Whether photography is needed and who arranges it
• What proof and credentials are available to use

Design and build

• Screen widths designs will be delivered at — more than one
• Whether a component system is included, or only page layouts
• Accessibility standard to be met, and when it is checked
• Performance expectations
• Who owns decisions the designs do not cover

Technical

• CMS, and who needs to be able to edit what
• Integrations required — CRM, booking, payments, analytics
• URL mapping and redirects, if replacing an existing site
• Who owns the domain, hosting and analytics accounts

The last item is worth confirming explicitly. Arrangements where a supplier holds the domain create genuine difficulty if the relationship ends.

Trends, and why they are not on this list

Nothing above is a trend. Clarity, hierarchy, readable type, obvious next steps and working on a phone were true a decade ago and will be true in another one.

Trends are worth knowing — they affect whether a site reads as current, which is a real consideration for some businesses. But they are a separate topic with a separate shelf life, covered in current design trends. Building a business site on trends alone produces something that will need redoing in two years for reasons unrelated to whether it works.

Frequently asked questions

What are the most important web design tips for a business website?

Decide what the site must accomplish before designing anything; say what you do in plain language above the fold; organise navigation around the words customers use; give every page one obvious next action; place specific, attributed proof beside the claims it supports; and make sure it works properly on a phone. Clarity outperforms visual sophistication consistently.

What are typical website design requirements?

The site's primary purpose and audience, the main action you want visitors to take, a page inventory with named content owners and dates, the screen widths designs will be delivered at, whether a component system is included, the accessibility standard, required integrations, and confirmation of who owns the domain, hosting and analytics.

How many pages should a business website have?

One page per distinct thing you sell, plus home, about, contact and any legal pages. A single page covering eight services ranks for none of them and answers none properly. Beyond that, add pages when there is something specific to say, not to appear substantial.

Should a business website have a homepage carousel?

Generally no. Carousels usually exist because a team could not agree what matters most, and they postpone that decision rather than resolving it. Most visitors never see the second slide, so anything important there is effectively unpublished.

How do I know if my website design is working?

Measure against the job you wrote down: enquiries, bookings, applications — whatever the site exists to produce. Supplement with observation, watching real people attempt the main task. Aesthetic opinions from colleagues are the least reliable input available, because they know what everything means already.

Where to start

Write one sentence describing what your site must accomplish. Then open your homepage on a phone and time how long it takes a stranger to understand what you do.

Those two exercises identify most of what is wrong with a business website, and neither requires a designer. If the sentence is hard to write, that is the finding — a site cannot communicate a purpose the business has not articulated, and no redesign will substitute for settling it.

When the answer is that the site needs rebuilding rather than adjusting, our website design services cover that work — starting with the purpose question rather than with layouts.