Rebuilding the homepage around four first-visit doubts

Why the homepage: it receives the most traffic and converts below the category benchmark, so a small lift there is worth more than a large lift anywhere else. I anchored the goal to two page-level metrics, conversion rate and bounce rate, and benchmarked against the Google search whole-year average rather than our own history.

Instrumented the page by section before redesigning it: click-through per section, drop-off between sections, and conversion from the pages each CTA leads to, so the new page could be read section by section instead of as one number.

Wrote the problem from the visitor's side, as a first-time visitor, in priority order: I am not clear you are a rental service or how you differ; you do not offer enough styles I like; I am hesitant to trust you; I cannot find my size or favorite brand. Each doubt became the owner of a section group and a hypothesis about which metric would move.

P1 went to the three sections that answer the top doubt, because they sit above the fold and are mostly copy. P2 went to the trust sections, which needed new design, brand assets and numbers we had to verify first. Page order follows doubt order, so a visitor meets the answers in the sequence they ask the questions.

Sized each section by what it needed: copy only, copy plus existing assets, or new design. Copy-only sections shipped first as a batch so one deploy could be measured against one baseline.

• Hero. From "Maximum Style. Minimum Effort." to "Menswear rental curated by expert stylists." One sentence that says what we are and who does it. Press logos kept, each linked to its article, sub-descriptions removed.
• How it works. Three steps, each one line: styling and curating, renting, swapping. One CTA to the full page.
• Why choose Taelor. Rewritten from features to end benefits, revised again on 12/28 after review: dedicated 1-on-1 stylist, 50+ years of expertise, new styles monthly with AI tips daily, no late or damage fees, own the style without buying.
• Trust. "300+ brands, 10,000+ items, sizes XS to XL" and a numerical claims block, with testimonials and a CTA after each.

Wrote the brief as a build spec: for every section, the pain point it answers, the goal, the copy, the CTA target (brand icons to Our Brands, inventory CTA to Membership, How It Works to its page), and the primary and secondary metrics. Added image alt text so search engines read each section, and a consistent tonality rule across all copy so the page reads as one voice.

primary metric per section: click-through · secondary: drop-off, conversion from the destination page

Because each section had its own click-through metric, a page-level win could be traced to the sections that produced it, and a section that did not move could be revised on its own without touching the rest. The How It Works "learn more" click-through and the conversion of visitors who arrived from it became the standing check on whether the rental model was landing.

Answer the visitor's first question first. The biggest lift came from the cheapest change: a hero that says "rental" and "stylists" in one line. Proof sections matter, but only after the visitor understands what is being proven.

A brief that names the metric per section is also the instrumentation plan and the QA checklist. Writing the three together once meant nobody argued afterwards about what "worked" meant.