TL;DR
- Most website conversations are about whether it looks right. That is a fair question — it is also the only one that ever gets asked properly.
- Looks decide whether you get considered. Everything after that is decided by clarity, speed and whether people can actually use the thing.
- Each of those has published evidence behind it, which means they can be argued about with something other than taste.
- Run the conversation in that order and design sign-off stops being a matter of who in the room has the strongest opinions.
Here is how the website conversation normally goes. Someone shares a link. Everyone looks. Someone says the header feels heavy, someone else prefers the old blue, and a decision gets made by whoever is most senior or most tired. Six months later nobody can say whether it worked, because nobody agreed at the start what working would look like.
The conversation you are having is not wrong — it is just first
It is tempting to open with “stop worrying about how it looks”. That would be bad advice. Stanford’s Persuasive Technology Lab published ten guidelines for web credibility, and the sixth is to design your site so that it looks professional or appropriate for its purpose — because, in the lab’s own words, people quickly evaluate a site by visual design alone.
So appearance is not vanity. It is the gate. What goes wrong is that it becomes the entire agenda, and the four conversations behind it never happen at all — even though those are the ones that determine whether the site earns anything.
The running order
- Does it look credible?The gate. Consistent spacing, real photography, no broken links, nothing that reads as careless. Necessary, and the fastest to fix.
- Is it clear?Can a stranger say what you do and who for, from the first screenful, without scrolling or interpreting?
- Is it quick?Not the homepage score on your own laptop. The experience on a mid-range phone on a normal connection.
- Can everyone use it?Contrast, labels, keyboard access. The part that is legally relevant and commercially invisible until it is not.
Notice that only the first is a matter of taste. The other three have measurements, and measurements are what turn a design meeting into a decision rather than a negotiation.
Where attention actually goes
The clarity conversation is usually settled by asserting things about scrolling. There is better evidence than that. In a 2018 eyetracking study drawing on more than 130,000 eye fixations from 120 participants, the Nielsen Norman Group found that users spent about 57% of their page-viewing time above the fold — down from 80% in its 2010 research, but still the majority.
Both halves of that matter. People scroll far more readily than they did, so the old panic about pushing anything below the fold is overdone. But more than half of all attention still lands in the first screenful, which makes it the most valuable space you own and the space most often spent on a stock photograph and a slogan.
If the most-viewed part of your website does not say what your business does, nothing further down will rescue it.
A worked example. A Brighton accountancy practice had a homepage headline reading “Clarity. Confidence. Growth.” above a photograph of a laptop. Replaced with “Accountants for owner-managed businesses in Sussex — fixed monthly fees, no surprises”, plus the phone number, in the same space. Same design, same brand, same everything else. The difference is that the second version answers the question a visitor arrived with.
The performance conversation has an actual price on it
Speed is where the design conversation most often gets overruled, usually by whoever wants the video header. It helps to know what the trade is worth. In Milliseconds Make Millions, research commissioned by Google and carried out by 55 and Deloitte across 37 brands, a 0.1 second improvement in mobile site speed moved 3.2% more retail shoppers from the listing to the product page and 9.1% more from the product page to the basket. Retail consumers spent 9.2% more and travel bookings rose 10%.
That is retail and travel data on mobile, so do not transplant it wholesale onto a professional services site — a buyer choosing a solicitor will tolerate a slow page in a way a shopper will not. What does carry across is the scale of the effect for the size of the change. A tenth of a second is not something anybody in a design meeting has ever consciously defended, and yet it is being spent, repeatedly, by sliders and font loaders and tag managers nobody has audited.
The conversation nobody wants to open
Then there is accessibility, which tends to arrive last and framed as a compliance risk. The scale of the problem is well documented: the WebAIM Million 2026 report found detected WCAG 2 failures on 95.9% of the top 1,000,000 home pages, averaging 56.1 errors per page, up from 94.8% the year before. Those are only the failures software can spot automatically, so the true figure is worse.
The reason to raise it in the design conversation rather than at the end is that almost all of it is decided by design choices: text colour on a background, whether a button says what it does, whether a form field has a label, whether motion can be turned off. Retrofitting those costs several times what deciding them correctly would have.
How to run the conversation
This is a change in meeting structure, not in budget. It works with the site you already have.
Four questions, in order
- Agree the job of the page before you look at it
One sentence: what should a visitor do, and what do they need to believe first? Written down, before the design is on screen.
- Test the first screenful on strangers
Five seconds, six people, one question: what does this business do and who for. Their words go on the wall.
- Bring a measurement to the speed argument
Field data from real visitors, not a lab score. Then decide what the header video is worth in those terms.
- Put accessibility in the design review
Contrast, labels and keyboard access reviewed alongside the layout, by the same people, at the same time.
Do that and the meeting changes character. Opinions about the blue still get aired, but they now sit alongside four things that can be checked, and the design that survives is the one that answers all of them rather than the one belonging to whoever spoke last.
Good to know
FAQs
Yes — it is what gets judged first, and Stanford’s web credibility work puts a professional appearance among its ten guidelines. The mistake is treating it as the only question rather than the first one.
Sort of. People scroll much more readily than they used to, but eyetracking research still puts the majority of viewing time above the fold. Treat it as the most valuable space rather than the only space.
Fast enough that nobody thinks about it, on a mid-range phone rather than your own laptop. Research into mobile retail found meaningful commercial gains from improvements of a fraction of a second, which is smaller than most people assume matters.
During the design review, not at the end. Most failures are design decisions — colour, labels, focus states, motion — and correcting them later costs several times what deciding them properly would have.
Agree what the page is for before anyone looks at it, then test the first impression on people who have never seen it. Once there is evidence in the room, the strongest opinion stops being the deciding factor.
Next step
Having the wrong conversation about your website?
We work with businesses across Brighton and London on the parts of a website that can actually be measured.