Homepage sections ranked by what actually moves conversion, not opinion
Homepage section order gets debated more than almost any other design decision, usually based on aesthetic instinct rather than actual data. Across the client rebuilds and tests we've run, a fairly consistent pattern emerges in what actually moves conversion versus what just feels important to include.
The hero's job is narrower than most people think
The hero section's job isn't to explain everything about the business, it's to answer "am I in the right place" within about three seconds and give a single clear next action. Heroes trying to communicate three value propositions and two calls to action at once consistently underperform heroes making one clear claim with one clear button, across every client test we've run.
Social proof works best placed early, not late
Client logos, testimonials, or trust badges are frequently placed near the bottom of the homepage as a final reassurance. Moving a condensed version of social proof higher, directly below the hero rather than after several sections of feature explanation, has outperformed the bottom placement in the majority of tests we've run, particularly for less well-known brands where trust is the primary early objection.
Feature lists convert worse than outcome statements
A section listing product features ("real-time sync, unlimited storage, team permissions") consistently underperforms a section restating the same information as outcomes ("never lose a file again, work with your whole team in one place"). The underlying information is often nearly identical; the framing around what the visitor actually gets is what moves the number.
FAQ sections placed before the final CTA, not after
Placing an FAQ section addressing common objections directly above the final call-to-action, rather than relegating it to a separate page, consistently reduces the "I have a question so I'll leave and think about it" drop-off we see in session recordings. It's a small structural change with an outsized effect on late-funnel hesitation.
What doesn't move the number much
Team photos, awards and certifications badges, and "our process" step-by-step sections tend to have minimal measurable effect on conversion in our testing, even though clients often want them prominently featured. They're not worthless, they build credibility for visitors who are already close to converting, but they're rarely worth prime homepage real estate over the sections that demonstrably do move the number.
Pricing sections: show it or hide it, but pick deliberately
Whether to show pricing directly on the homepage versus behind a "contact us" or a dedicated pricing page depends heavily on the buying context, and we've seen tests go both ways depending on the business. For simple, self-serve products, showing pricing upfront tends to filter out unqualified visitors early and speeds up the decision for qualified ones. For complex or highly customized offerings, an upfront number without context can scare off visitors who would have converted with a bit more framing first.
The mistake we see most often isn't picking the wrong option, it's not picking deliberately at all, defaulting to whichever pattern a template shipped with rather than testing what actually fits the specific offer and price point. A quick A/B test on this single element, visible pricing versus a gated pricing page, is one of the highest-signal, lowest-effort tests available to most homepage owners.
Whether to show pricing directly on the homepage versus behind a "contact us" or a dedicated pricing page depends heavily on the buying context, and we've seen tests go both ways depending on the business.
How long a section actually needs to be tested before trusting it
A homepage section change needs enough traffic and enough time to produce a statistically meaningful result, and reading a test after only a few days of data is one of the most common ways teams draw the wrong conclusion. We generally recommend running a homepage section test for at least two full weeks, capturing both weekday and weekend traffic patterns, and for lower-traffic sites, considerably longer than that before treating any single test result as conclusive rather than noise.
Video versus static hero, a test we run often
Clients frequently ask whether a background video in the hero section helps or hurts conversion, and our testing shows the answer depends heavily on execution rather than the format itself. A slow-loading video that delays the hero's headline and call-to-action from appearing quickly tends to hurt conversion even when the video content itself is compelling, since the visitor is waiting rather than acting. A well-optimized video that loads without blocking the text and button above it can perform as well as or better than a static image, but the bar for "well-optimized" is higher than most teams initially plan for.
We generally recommend testing a static hero first, since it's cheaper to build and easier to get right, and only moving to video once there's a specific, compelling case for it and the engineering budget to implement it without compromising load time. A mediocre video hero that slows the page down is a worse outcome than a well-executed static one, even though video often feels like the more modern, ambitious choice at the proposal stage.
How mobile changes this entire ranking
Everything above assumes a desktop-first reading of the homepage, but a majority of traffic for most of our clients arrives on mobile, where screen real estate is far more constrained and every section competes harder for scroll depth. On mobile specifically, we've found the social proof section benefits even more from an early placement than it does on desktop, since a mobile visitor scrolling past the hero has already made a bigger commitment of attention than a desktop visitor glancing at a wider screen, and that attention needs to be rewarded with reassurance sooner rather than later.
How we present these findings to a client with strong opinions
Founders and marketing leads often have a genuine, sometimes strongly held preference for a specific homepage layout, usually shaped by whatever design trend or competitor site inspired them most recently. Rather than arguing the point on opinion, we frame testing results as evidence gathered specifically for their own audience and offer, not a universal rule imported from someone else's business. That framing tends to land better than simply asserting our recommendation is correct, since it invites the client into the process of confirming or overturning it with real data from their own site, rather than asking them to simply defer to our judgment.