The most common UX complaint from startup founders is "the site is beautiful but it does not convert". The two statements are not in conflict. Visual polish and conversion design are different skills. A site can win design awards and lose customers in the same week. Aesthetics handle the first impression. Usability handles whether the visitor stays long enough to convert.
This is the seven-mistake list we audit every SARVAYA conversion engagement against. Most underperforming sites carry four or five of these. Fixing them rarely requires a redesign. The work is usually one to two weeks of focused UX edits on the existing build.
The gap between beautiful design and usable design
Award-winning sites optimise for the first three seconds. Conversion-driven sites optimise for the eighth second through the final form submission. Both matter. The mistake is treating them as the same skill.
The first three seconds are visual. Does this look credible? Does it match the price point? Does it feel like a brand I want to associate with? The eighth second is structural. Can I find what I came for? Is the CTA where I expected it? Does the page respond when I tap? Sites that fail the second test lose conversions even when they pass the first.
The seven UI/UX mistakes we see on almost every audit
Each of these is silent. No visitor calls support to complain about it. Each one shaves a few percent off the conversion rate. Fixing four of seven typically lifts conversion 30-50% on the same traffic.
- Unclear primary CTA. Three buttons of equal weight in the hero section. The visitor cannot tell which is the intended action and picks none. Fix: one primary button, distinct colour and size, repeated in every viewport.
- Slow first interaction. The page looks loaded but tap-to-response latency is 800ms because JS is still parsing. The visitor taps twice, fires two events, and bounces. Fix: target INP under 200ms by deferring non-critical JS.
- Decorative micro-animations on every element. Buttons that ripple, cards that lift on hover, headlines that fade in word-by-word. Each one is fine in isolation. Stacked together they make the page feel sluggish and pretentious. Fix: pick three intentional animations and remove the rest.
- Broken mobile layouts on edge cases. The hero looks fine on a Pixel 7. On a 360px Android device or a foldable, the headline wraps oddly and the CTA disappears below the fold. Fix: test on three real devices including the smallest viewport you support.
- Hidden contact paths. The only way to talk to the company is to fill out a 12-field form on the contact page. No phone number, no WhatsApp, no email. Fix: three contact paths visible from every page, in priority order for your audience.
- Inconsistent component styling. Three button styles across the site. Two card styles. Headings shift weight between pages. Each one signals to the visitor that the brand is improvised. Fix: a small design system locked at five components.
- Excessive form fields. 12 fields on the lead form. Most are not used by sales. The drop-off between field one and field twelve is 60-80%. Fix: three required fields, optional fields below a "more details" disclosure.
How cognitive load loses visitors in the first eight seconds
Cognitive load is the mental effort the visitor spends understanding the page. Every element competes for attention. Every animation costs decoding time. Every paragraph of marketing copy that does not directly answer "what does this do for me" is friction.
The eight-second rule comes from Microsoft's attention span research and matches what we see in heatmaps. If the visitor cannot answer three questions in the first eight seconds (what is this, what does it cost, can I trust it), they leave. The fix is not more design polish. It is fewer competing elements and a clearer hierarchy of information.
The best UX edit on most underperforming sites is removal. Cut the third CTA. Cut the rotating hero carousel. Cut the testimonial slider. Every removal raises clarity faster than any addition.
What a UX audit actually covers
A formal UX audit takes a working site and produces a ranked list of fixes with expected conversion impact. The structure is identical across industries.
- Heatmap and session recording review. Two weeks of real-user data via Hotjar or similar. Identifies dead zones, rage clicks, and dropoff points.
- Five-second test on the home page. Five strangers see the page for five seconds and report what they think the site is for. If three of five get it wrong, the hero section is failing its job.
- Mobile device testing on real hardware. Not Chrome DevTools emulation. Three real Android devices and one iPhone, including a smaller screen.
- Form analytics with field-level dropoff. Where exactly the visitor abandons the form. Almost always one specific field is responsible for most of the loss.
- Speed audit with real-user metrics. Lighthouse synthetic plus PageSpeed Insights field data. Both views matter because they measure different things.
The difference between a UI refresh and a UX overhaul
A UI refresh updates the visual layer: new fonts, new colours, new component styling. It is a two to four week project and addresses the "the site looks dated" complaint. It does not address conversion structure. A UX overhaul restructures the information architecture: page flow, CTA hierarchy, form design, content priority. It is a six to ten week project and addresses the "we get traffic but no leads" complaint.
Most founders ask for a UI refresh when they need a UX overhaul. The visual layer feels stale because the structural layer is wrong, and a new coat of paint on a broken structure still feels broken. The cheaper move is to fix the UX first and then judge whether the UI also needs work. According to Nielsen Norman Group's usability research, the conversion impact of UX-level fixes outweighs visual updates by roughly 3:1 in measured A/B tests. The Google web usability guidelines include the same prioritisation in their Core Web Vitals framework: interaction latency outranks visual polish for ranking signals.
How we run UX audits at SARVAYA
Every conversion engagement starts with a one-week audit producing a ranked fix list. Most lists have 12-20 items. Top five usually account for 80% of the expected conversion lift. We ship those five inside the first sprint and re-measure conversion at week three.
Our web development service includes UX audits and remediation for client sites. For the deeper conversion structure that pairs with the UX work, see our organic lead generation playbook. The role of designers in the broader project context is covered in our UX designer importance piece. Talk to us at SARVAYA if you want a one-week UX audit scoped against your current site.
