The 5 Website Principles Most Companies Get Wrong

Some of the best-looking websites we seen or audited were quietly failing the businesses that paid for them. They won design awards. They documented well for the case study, the client loved the site for a year. But the site just did not convert, and nobody could say exactly why.

The signs your website is outdated are rarely the obvious ones. A dated colour palette, an old logo, or an off-brand look is easy to spot. The expensive problems sit underneath: unclear goals, broken structure, slow load times, lapsed security that puts customer data at risk, copy written for the wrong person. Google and other search engines notice some of these before your potential customers do, and both leave quietly. These are the fundamentals, and they get skipped because they are invisible until something breaks.

Across the projects we run at Tennis, a UX design firm focused on measurable outcomes, the same five principles come up again and again, whether the business is a professional services firm or a product company. Here is where most companies get them wrong.

Clarity beats cleverness

Designers love a high concept. So do the people who hire them. The trouble is that clever rarely survives contact with a real business goal.

A website can look like nothing else in its market and still fail, because looking different was the only brief it was given. Before anyone opens a design tool, someone has to answer a plain question: what is this website for, and how will we know if it worked? If the goal is reach and brand awareness, build for that. If the goal is qualified leads for your services, that is a different website with different features, mapped to a real customer journey. The mistake is launching something beautiful and only then asking what it was supposed to do.

This is mostly a documentation problem. Most businesses we work with have never written down the goal of their company website in a way the whole team agrees on. There is no KPI to point at, so project teams fall back on "it looks great" as the measure of success by default. That is how you end up with a website that earns compliments and no pipeline.

First impressions are unforgiving. Research by Lindgaard and colleagues found that users form a reliable judgment of a page's visual appeal in about 50 milliseconds, and that first look shapes whether they trust you before they have read a word. Clarity is what turns that snap judgment into something useful. A visitor should grasp what you do, who you do it for, and what to do next before they have decided whether to stay.

Write the business goal down. Attach a number to it. Then let the design serve that, and not the other way around.

Structure is the real UX

Ask most people to picture user experience and they picture screens, or a slick user interface. The real work happens before any screen exists, in how the website is organized: what lives where, what connects to what, and the order someone moves through it. Understanding how UX, CX, and UI fit together makes it easier to see why structure matters as much as visuals.

Structure gets undervalued because it is invisible. You cannot see information architecture the way you can see a hero image and say "make it blue." It shows up as diagrams, sitemaps, content models, and strategy conversations, which already asks a lot of a busy team. So people skip ahead to the part they can react to: the aesthetics.

The catch is that you do not notice structure until it is bad. A site with no architecture feels like a building with no floor plan. Website visitors cannot find what they came for, and confusing navigation quietly kills conversions long before anyone files a complaint. UX surveys commonly cited in e-commerce research suggest that around 88% of users are less likely to return to a site after a bad experience, and many of those experiences are structural rather than visual.

There is a line we keep coming back to, from a senior designer at a business you would know: the difference between art and design is that art makes you think, and design makes it so you do not have to. A well-structured site does the thinking for the visitor. They never have to work out where to go, because the path is already laid out across all the pages.

You would not build a house without a plan. Start with the map. The visual design and the wider user experience have somewhere to live once the structure is sound. If the line between this and the visual layer still feels fuzzy, we pulled it apart in our piece on UX versus web design.

Speed is a design decision

Speed usually gets filed under engineering. It belongs to design too, because most of what slows a site down is a design choice: the animation library, the high-resolution hero video, the third typeface, the carousel nobody asked for. In most web projects, those decisions live inside a bigger web development process that has to balance performance with everything else you are trying to ship.

There are two kinds of speed worth caring about. Real speed is what a stopwatch measures. Perceived speed is how fast it feels. You manage both. The benchmark most teams know is the three-second rule. Google's mobile research with SOASTA found that about 53% of mobile visitors abandon a page that takes longer than three seconds to load. The stakes keep climbing, since recent traffic analyses estimate that around 63% of Google's organic search visits in the US come from mobile devices, often on a worse connection than your office wifi. Google and other search engines factor mobile performance and basic security like SSL into rankings, so a slow, unsecured, or non mobile friendly site loses ground in search results before a human ever judges it.

When you genuinely cannot make something load faster, you can still manage how the wait feels. A progress indicator or a light loading animation tells the visitor something is happening and buys a second of patience. What you cannot do is ignore the problem. We recently looked at a B2B site where authentication alone took six seconds before the real content appeared. Six seconds of staring at a blank state, wondering whether it is broken, your connection, or you.

Every heavy asset is a tradeoff. Nice things have weight, and that weight has a cost in attention and rankings. Left unchecked, it is the kind of slow-accumulating problem we wrote about in website technical debt. If something is going to look impressive, the speed box has to stay checked.

Your website has to work for someone who knows nothing about you

This is the principle we see violated most. Your website is not for your own team, and it is not for the customers you already have. It is for potential customers who have never heard of you and have about ten seconds of patience.

Getting this right starts with work that has nothing to do with design. You have to know who your ideal customer is, what they care about, and the language they use to describe their own problem. Most businesses have not done that work, so the website ends up speaking to the founder, the board, or whoever held the strongest opinion in the kickoff. None of them are the visitor. In a lot of businesses that work now sits with a small, cross functional team wearing many hats, with nobody clearly owning copy, or a clear website project brief that defines scope and goals.

Then there is the copy problem, which has gotten worse. What used to be "our content is not ready" is now "we will just have AI create it." We understand the temptation. The risk is that generic copy does not know your ICP, reads like every other site in your market, and ignores how people actually consume content on the web. Visitors scan before they read. They scroll, catch on a heading, and only then read two or three short paragraphs. Copy decks that arrive as dense, essay-length blocks fight the design and lose. AI can get you partway, but it cannot do the thinking about who you are talking to. We unpacked that shift in AI, UX, and marketing.

A practical habit: revisit your core copy every six months against your ICP. You do not need a full-time copywriter on staff. You need someone, once or twice a year, to check that the words still match the person you are trying to reach, reflect your brand, and carry the unique value proposition only your business can claim.

Every page needs one job

One page, one job. One topic, one primary action, one reason to exist. This keeps the experience focused, and it keeps your business from competing with itself, where two pages chasing the same keyword cannibalize each other in search engine results.

This is also where people misread the rule. That does not mean one call to action everywhere. The clearest example is the homepage. When someone lands there, the site often does not yet know who they are or why they came. So the homepage has a single role, and it is a routing one: answer who you are, what you do, and who you do it for, quickly, then send each visitor down the right path. Think of it as a traffic mediator. A few clear sections that preview the main navigation, each with a CTA that branches to the page built for that intent, the way we approach website design and development for marketing teams.

A SaaS homepage can get away with a single "book a demo," because the whole site exists to drive one conversion. B2B is messier. Different visitors arrive with different questions, different roles, and different levels of readiness, and the homepage has to acknowledge that without trying to close all of them on the spot. The same logic applies to contact forms and any other primary action: match it to what that specific page is for.

The principle holds either way. Decide the purpose of a page before you design it, the same way a good product designer scopes a single feature instead of ten. For deep pages, that purpose is usually one action. For the homepage, it is sending the right person to the right place.

Where to start: auditing the signs your website is outdated

If your website is not performing, resist the urge to jump straight to a redesign. Start with an audit instead, using templates and audit resources for web and marketing teams.

Look at your architecture and your analytics data first, and ask one question: does it match your ICP? Then name the single primary goal of the website. Most teams carry many goals and never rank them, which is exactly how a project loses focus. It is the same exercise we run in our workshops, where we sort audiences into primary, secondary, and tertiary, the approach behind our HOLO framework. Priorities force clarity. Do the same with your pages and your business goals.

Fix the structure and the primary goal, and most of the other problems on this list get easier to solve. The signs your website is outdated almost always trace back to one of these five fundamentals, not to the colour of a button. For more on how design practice is changing, we unpacked signals from Config in Design is the Process, and if you want the longer version of turning the website into a real business asset that supports your services and your position in the market, we covered it in B2B website strategy and in our broader blog and resources on digital strategy and UX.

The businesses that get websites wrong rarely have bad taste. They just never decided what the website was for.