Most of the advice businesses receive about web design tends to start and end with the visual aspects of the site.

But a surprising amount of what makes a website feel consistent comes down to details that cost almost nothing to fix. They’re the sort of thing that experienced developers do without even thinking about – so automatically that it never appears on a quote, never gets discussed in your meetings, and never gets checked when it’s missing.

Here’s the main things I check when opening a site for the first time.

Website addresses should always be consistent

Type your site into a browser and try each of these:

• example.com
• www.example.com
• http://example.com
• http://www.example.com

All four should land on exactly the same address, with the padlock showing. If they don’t, you effectively have several versions of your website running at the same time.

That matters for three reasons:

• Search engines can treat different URLs as separate sites and split the credit between them
• Website analytics or tracking can end up recording visits inconsistently
• Links people share can point to a version that isn’t secure, which shows a warning in the browser before anyone sees a single word on the website

You need to pick one canonical domain and redirect everything else to it permanently.

It genuinely doesn’t matter which is chosen, as long as you are consistent and stick with it:

• https://example.com
• https://example.com/
• https://www.example.com
• https://www.example.com/

With the www subdomain or not, and with a trailing slash or not – but never both.

And it doesn’t stop at the home page.

Website internal links need to be consistent, too

Choosing a home page style only solves one of your problems – you also need to consider how every internal page looks.

Here are seven URL segments that could all serve the same services page:

• /services
• /services/
• /Services
• /Services/
• /services.html
• /services.php
• /index.php?page=services

Each of these can be treated by search engines as separate pages – even though they are all pointing to identical content – and every instance is a version a search engine has to work out how to treat.

The solution is simple: pick a format and enforce it.

Decide whether your website uses trailing slashes or not, force everything to lowercase, and redirect the alternatives to the correct version.

Avoid file extensions in web page URLs

File extensions deserve a specific mention: if your pages end in .html, .php, .asp or similar, your addresses are advertising the technology with which the site was built. That may not be a problem today, but the day you change technology either every link on the site breaks, or the .php address extension must be carried to a site that no longer runs PHP.

Extensionless URLs (such as /services or /services/ rather than /services.html or /services.php) outlive the platform they were built on.

If you already have extensions in place, don’t just strip them and move on – redirect the old addresses to the new ones so the links and rankings automatically follow across.

The related habit is to keep your website’s URLs short and readable. For instance, /services/website-speed/ is easier to say over the phone, easier to scan in a search result, easier for a visitor to understand at a glance, and easier to link to than /index.php?category=7&page_id=43.

When separating words in URLs, use hyphens – not underscores

Consider the following URLs:

• example.com/about-us/
• example.com/about_us/

To a human, these are as readable as each other – they both are clearly linking to the “About Us” page – they might even be considered interchangeable.

But Google explicitly recommends using hyphens instead of underscores ¹.

For historical reasons, many search engines will consider about_us to be equivalent to aboutus, treating it as one word. And a website should not try to rank for the word “Aboutus” because it’s not a real word.

This gets worse the longer the phrase, so website_performance_tips is seen as websiteperformancetips.

That makes it harder for search engines to actually understand what the page is supposed to be about – and if they struggle to make sense of what the URL represents, that can hold back the page from ranking well.

Avoid combining words in URL segments, but it can be done in domain names

Related to the point above, you should never combine words in your URL segments.

If you’re unsure about the difference between a URL segment and domain, consider the following URL:

• perthlandscapingandgardening.com.au/our-services/backyard-design/

In this instance, the domain name is perthlandscapingandgardening.com.au, the first segment is our-services and the second segment is backyard-design.

Search engines understand that a lot of domain names will contain multiple words, so they use different rules to consider the words within the domain – in the above instance, you will be able to search for “Perth landscaping and gardening” and it will be able to recognise it as 4 separate words in the domain name.

However, the same rules do not apply to the URL segments – without hyphens to separate the words, they’re treated as a single word.

So if instead, the URL was perthlandscapingandgardening.com.au/ourservices/backyarddesign/, then search engines will think you’re trying to rank for the words ourservices and backyarddesign – which are not words, and not what people type into search engines when they’re looking for a landscaper / gardener to design their backyard.

If your web designer / developer wants to be minimalist in your URL structure, you should understand that there is a cost – and that cost is usually your positioning within search engines, and ultimately the customers that come from that.

Header and footer components should be consistent

These elements should be identical regardless of what web page you’re currently visiting.

There are some exceptions, which we’ll go through below, but essentially these should be treated as landmarks where visitors can always easily return to the pages they will mostly need.

And as they are positioned at the top and bottom of each page, they are the first and often last thing your visitors will see when interacting with your web page, so their importance cannot be overstated.

The header logo should link to the home page

This one is a universal convention, but should be mentioned: people know to click the logo to return to the main page. It’s been standard for over twenty years now, and people do it without consciously deciding to.

When the logo isn’t a link, or if it is a link on all pages except for one, you probably won’t get a complaint sent through your contact form – most likely they will quietly tap the back button a few times or leave.

From my experience, there is only one exception to this rule, and that’s if there is a standalone sub-section of a website – for instance, if there is an online store within a firm’s website, or a non-profit structure that exists within a for-profit website. In these instances, there still needs to be consistency, but there needs to be a decision about whether the logo link goes to the internal sub-section’s or parent structure’s home page, or a combination of the two.

The main menu is not a sitemap

I often see these two get treated as the same thing, but they’re not.

A website’s main menu is an editorial decision – generally, it’s the primary routes that most visitors need most of the time, kept short enough to scan within a couple of seconds. A header menu with dozens of items isn’t thorough, it’s a wall of text – one that many people will simply ignore.

On the other hand, a sitemap is a link to everything published on a website. Every service page, article, case study and landing page – including ones from campaigns that were run several years ago and forgotten about.

A lot of business website owners assume that if a page isn’t in the menu, nobody will ever see it. But in practice, people arrive at unlinked pages all the time:

• From search. Search engines index pages, not menus. Someone searching for a specific problem lands directly on the page that answers it, often several layers deep, without ever seeing the home page.
• From links inside web page contents. An article that links to a detailed guide sends people to a page that may not appear in the navigation at all.
• From links outside of the website. Emails, directory listings, social media posts, old newsletters, other businesses linking to specific pages – your organisation cannot control what other people link to.
• From people typing a URL to see what happens. It’s more common than you might think – someone tries /pricing or /careers on the off-chance, and if something loads, they now have direct access to something you probably never thought they would see.

Three things follow from that:

1. Every page in a website needs to work as an entry point. If a visitor’s first impression of your business is a page halfway down the site, that page has to tell them who you are, what you do, and where to go next. Pages written on the assumption that everyone arrives from the front page can lose some of your highest-intent prospects.
2. Taking a link out of the menu doesn’t make the page invisible. Old campaign pages, superseded service pages, thank you pages, stray PDFs uploaded years ago – they’re all still live, still indexable, and still capable of showing someone information you’d rather they didn’t see. If a page shouldn’t exist anymore, remove it and redirect it. Hiding it from the menu isn’t removing it.
3. Pages with no links pointing at them from anywhere are hard for search engines to find and tend to perform poorly as a result. If a page matters, something on your site should link to it – a related article, a services index, or a link in the footer works. There are also self-updating sitemap pages that can be easily added.

A website’s footer is the usual pressure valve here. It’s where the secondary pages live – policies, detailed sub-services, careers, sitemap – so the main menu can stay short without anything becoming unreachable.

Menu item links should show people where they are

Similar to a map with a “you are here” marker so that people can understand the context of what they’re looking at, your website should also have an indicator showing the current page they’re on.

It doesn’t need to be drastically different – a heavier font weight, an underline, a background or text colour change – but it should be obvious at a glance.

It should also have specific markup so that screen readers and AI crawlers know which page is current – not just sighted visitors. In practice this means adding aria-current="page" to the active menu item, on top of the chosen visual treatment.

Breadcrumbs do the same job on sites with nested pages. If someone lands on a service page from a search result, the breadcrumbs tell them immediately what section they’re in and give them a simple path back up to the rest of the services. It also gives search engines a clean signal about how a website is structured.

Links should represent what happens when they’re clicked

Links are a very important aspect of how search engines understand the structure of, and rank, web pages ².

Link text should make sense on its own. Consider the following link text variations:

• Click here
• Read more
• Read the guide
• Read the website accessibility guide

Imagining that these all link to the same page – which snippet of text provides the most information about what you might get after you click on it? It should be obvious that “Read the website accessibility guide” as a standalone link makes the most sense about what could be on the other side.

Usually, a “click here” or “read more” link is surrounded by context which is easily understood a human – but search engine crawlers often view a website from the perspective of its links, which doesn’t take the words around the links into consideration. From that perspective, a phrase like “learn more” doesn’t describe what is on the other side of the link, and it thinks your website is trying to rank for that exact term.

Links should take advantage of all five visual states, not just one or two

Most website owners don’t know that browsers allow us to create multiple different visual states for each button, because most sites only style one of them properly.

Idle is the default state of a button, and what your design agency will always present when showing how the website looks.

Hover is what happens when a mouse cursor intersects with the link. This is usually done automatically by web design agencies, but it doesn’t exist on a touchscreen device like a phone or tablet, so it shouldn’t be the only feedback offered.

Focus is what happens when someone reaches an element using the keyboard, usually with the Tab key. By default, to help improve accessibility, browsers add a visible outline – but it’s remarkably common for web designers to deliberately remove that indicator because someone didn’t like how it looked. However, that makes the site genuinely unusable for anyone navigating without a mouse, which often includes people with motor impairments. If the default outline doesn’t suit your website’s design, it should be replaced with something that does, not with nothing.

Active is the pressed state – the moment between a click and the link’s action taking place (which is usually page navigation). A small visual change here makes the interface feel responsive rather than laggy, especially on slower connections where there’s a real gap before anything happens. You’ve probably experienced it – clicking a button and not knowing if it actually registered the click or not.

Visited is a different visual state shown when you’ve already been to a page. It used to be standard practice, and is built into the browser’s default link styles to show as purple instead of blue, but this is usually overridden by web designers who think that consistent colours are more important. However, it’s genuinely useful navigation information: on a long list of articles, resources or case studies, a :visited style tells returning readers exactly what they’ve already seen. It doesn’t need to be the same default purple colour – a slight change in shade is enough to be informative without being ugly.

Note: for security reasons, there are limitations to what styles can be applied to visited links – generally only colour can be changed, not things like font size or padding.

A website should work without scripts

One core aspect of modern web development is the use of in-page scripts, which allows web developers to dynamically change and update contents based on certain events or conditions.

For instance, we might build a contact form that only shows some of its fields when certain checkboxes have been selected – this generally requires some code that is actively running on the page, watching for changes, and then reacting to those changes when they’ve been made.

But scripts can also be easily overused, and behind images they’re often the greatest source of slowness in a website.

For the best possible website performance, scripts should only ever be invoked as a last resort – if it’s not possible to achieve the same result through pure markup or styling.

And for the 1 ~ 2% of website visitors that disable scripts, the website should still work perfectly ³.

A website should make sense without styling

This tends to surprise a lot of people, including web designers – but the web was initially built as a document sharing platform, and all web pages today still inherit from that basis.

Search crawlers, AI assistants, screen readers, reader modes and email previews don’t see your website the way most users see it. They read the underlying structure (the headings, paragraphs, lists, links and labels) and essentially ignore the visual layer sitting on top.

So if you turn the visual styling off on your website, is it possible to understand what the pages are about? This is possible by disabling CSS in the browser’s developer tools or by using a reader mode. What you want to see is a document that still reads as a sensible page: one main heading, appropriate subheadings in a logical order, content in the order it should be read, and links that describe where they go.

But what we often see is a jumble:

• A humungous wall of links at the top of the page rather than just what’s required
• Headings chosen for their visual size rather than their meaning
• Important text baked into images
• Content that only makes sense once it’s positioned visually

If a page falls apart without its styles, it’s more difficult for search engines and AI tools to understand and summarise accurately – and that is exactly how people find business websites.

Other foundational aspects of a business website

These don’t require a full section of their own, but collectively they make a real difference.

• Every page needs a unique, descriptive title. The browser tab and search engine results pages both use it.
• Convert phone numbers and email addresses into links. If it looks clickable, it should be clickable.
• Use appropriate form field types. Email address inputs should use type="email" rather than generic text, and phone fields should show a number pad instead of the regular keyboard.
• Make sure clickable links are big enough. Small links crammed together in a footer are frustrating when navigating on a phone.
• Add alt text to images that carry meaning. For purely decorative images, this should be left empty.
• Redirect old URLs when pages move or get deleted. Every broken link is a lost visitor and search engine ranking.
• Make your 404 page useful. Don’t just apologise – provide a search bar, access to prominent links, contact details.
  • Depending on your brand, adding some humour to this page can also help.
• Add a skip link at the very top of each page – these tend to be hidden until receiving keyboard focus – that jumps straight past the header and menu to the main content.
  • Without this, anyone navigating by keyboard or screen reader has to tab through every menu item before reaching the page they came for, on every single page. It’s only a few lines of code but is one of the highest-value accessibility fixes available.
• Don’t force links to open in new tabs unless there’s a good reason – and there’s usually never a good reason. If a user wants to open a link in a new tab, they know how – but when this is forced on them, it not only breaks the back button, but their expectations and sense of control are violated.
• Set the website’s page language. It’s one HTML attribute, and it tells screen readers and translation tools how to pronounce and process your content ⁴.

Why website fundamentals are worth your attention

Individually, none of these will drastically transform your enquiry numbers – but collectively it provides the foundations for a smooth and solid website experience.

A beautiful web design sitting on top of inconsistent URLs, outdated page structures and poorly-written links is like a house with lovely paintwork and a cracked slab.

Fixing the basics doesn’t just remove small irritations – it makes everything done afterwards work better, whether that’s a paid marketing campaign, a sitewide redesign or running multiple conversion rate tests.

The good news is that most of these can be checked in about 15 minutes with nothing but a browser, and most of them can be fixed in an afternoon.

If you’d like a fresh set of eyes on your website, request a website review and I’ll check everything myself and tell you what I find.

1. Google’s guide to URLs: https://developers.google.com/search/docs/crawling-indexing/url-structure#use-hyphens-to-separate-words ↩︎
2. Google’s guide to website links: https://developers.google.com/search/docs/crawling-indexing/links-crawlable ↩︎
3. Developers should do what is called “progressive enhancement” where the baseline website works well without scripts, but is enhanced when they’re enabled ↩︎
4. For Australian business websites, this should be <html lang="en-AU"> ↩︎