Every fortnight I review a CRO audit somebody on the team has put together. The pattern is consistent. The biggest lift hides in the boring stuff, not in clever experiments. Here are the five fundamentals we audit on every CRO retainer before anyone gets to draw a fancy variant.
1. The headline above the fold
Headlines are still the single biggest lever on a page. The one we replace most often is "Welcome to [Brand]" — a phrase that does no work for the visitor. A useful headline says what the product is, who it is for, and what changes. Three of our last four retainer wins came from a headline rewrite alone.
2. The primary CTA — count it, name it, repeat it
A page should have one primary action. Count yours. If you have three "primary" buttons fighting for attention, the user picks none. Name the CTA in the language of the visitor ("Get a quote" beats "Submit"), and repeat it sensibly at the bottom of the page after you have made your case.
3. The proof block, treated with respect
Customers want proof. Most sites bury it. Move your strongest testimonial, case study stat or logo wall directly underneath your hero block. Not below the fourth marketing section. Not in a separate /testimonials page nobody reads.
4. Form fields, ruthlessly trimmed
Each field on a form is a cost. We audit forms by asking: "what happens if we remove this?" Phone number is the most over-requested field — we have killed it from a contact form on five separate projects with zero downside.
5. Loading speed, especially on mobile
Lighthouse mobile score below 70 will hurt conversion before any of the above can help. The fix is rarely glamorous: compress images, defer scripts, kill the cookie banner that fires twelve trackers. We track this as a quality gate at the end of every sprint.
The boring experiments that win
If you are running a CRO programme, here are the experiment types we have the highest win rate on, in order:
- Headline rewrites against the same hero
- CTA label rewrites
- Form field reductions
- Moving proof higher up the page
- Removing a homepage slider that nobody asked for
The interesting variants — radical layout shifts, scroll-driven animation, AI-personalised hero copy — are fun, but they win maybe one experiment in ten. The boring stuff wins one in three.
What this means for new builds
If you are starting a new website project, build these fundamentals in by default. You will skip a whole quarter of remedial CRO experiments later. The right time to design for conversion is at design phase, not the quarter after launch.
— Prashant leads the CRO and analytics practice at Vapordex Studio.