What Are Companies A/B Testing on Their Hero Sections Right Now?

The hero is the single most-tested area: 488 of 2,160 annotations (23.2%) across 405 distinct experiments Lazyweb Research observed by July 2026. It splits almost evenly by platform — 272 web and 216 mobile — making it the only major area tested heavily on both. Average model-assigned impact is 3.28/5, mid-pack, with 28.1% of hero tests scored 4+/5. Observed patterns center on headline repositioning and swapping passive heroes for working tools. [1]

The hero is tested more than any other page area — 488 annotations (23.2% of 2,160), split 272 web / 216 mobile — Lazyweb Research, July 2026.

Lazyweb Research · n=488 · Published 2026-07-07

landing-pageexperimentsux-patternswebmobiledesign

How much, and where

Hero testing at a glance: [2]

MetricValue
Annotations488 (23.2% of 2,160)
Distinct experiments405
Web / mobile split272 / 216
Avg impact (1–5)3.28
Share scored 4+/528.1% (137/488)

Unlike offer, pricing, and header — which are heavily mobile/paywall — the hero is the one area tested at scale on both web and mobile, so it's the most cross-platform bet a growth team makes. [3]

Observed patterns

From the highest-impact hero experiments Lazyweb Research logged (all model-scored impact 5): [4]

  • Swap a routing decision for a working tool. Wise replaced two chooser cards ('Money Transfer Comparison' / 'Travel Money') with a live GBP→EUR comparison widget and an above-the-fold provider table — rate-searchers get the answer instead of a second click.
  • Narrow the headline to the real ICP. Replo repositioned from 'Sell Anything with Replo' to 'Conversion rate optimization for serious Shopify teams' and dropped the AI prompt box for standard signup/demo buttons.
  • Put commerce above editorial. SSENSE replaced an editorial-magazine fold with a 'SHOP FW25 NOW' seasonal campaign and gendered entry points, betting repeat traffic is shoppers, not readers.

The through-line: teams trade broad, passive heroes for something narrower and more actionable. [4]

How to apply it and caveats

If your hero routes users to a choice, test collapsing that choice into an inline tool or a single narrowed headline — that's the recurring high-impact move. Because hero impact is only mid-pack (3.28 avg), don't assume a hero test is your highest-leverage slot; the offer and pricing areas score higher. Caveat: these are observed diffs with LLM-inferred rationale and model-assigned impact, not measured lift. [5] Named examples are individual observations, not a representative sample.

The numbers

StatComputed from
Hero 488 annotations (23.2%), 405 experiments, 272 web / 216 mobile, avg impact 3.28statpack area_HERO
Hero: 488 annotations, 405 experiments, avg impact 3.28, 28.1% scored 4+ (137/488)statpack area_HERO + high_impact_share_by_area
Hero splits 272 web / 216 mobile; offer/pricing/header are mostly mobilestatpack platform_area_split
Hero impact-5 examples: Wise (widget), Replo (headline), SSENSE (commerce fold)statpack qualitative HERO entries
Experiments are observed diffs with LLM-inferred rationale and model-assigned impact, not measured liftstatpack methodology note
Methodology. Universe: 488 hero annotations across 405 detected experiments (subset of 2,160 annotations / 1,126 experiments), July 2026, labelled and impact-scored by LLM from paired screenshots. Named companies are single observations; impact is relative model signal, not measured lift.

Sources & citations

  1. [1] Lazyweb Research analysis of 488 hero annotations (405 experiments) within 2,160 total, July 2026. HERO is the most-annotated area; 272 web / 216 mobile.
  2. [2] Lazyweb Research analysis of 488 hero annotations, July 2026. Impact avg 3.28; 137 of 488 scored 4+/5.
  3. [3] Lazyweb Research analysis of 2,160 area annotations, July 2026. Platform × area split; hero is the only major area tested heavily on both web and mobile.
  4. [4] Lazyweb Research qualitative review of top-impact hero experiments, July 2026. Wise, Replo, SSENSE — individual observations scored impact 5 by the model.
  5. [5] Lazyweb Research methodology note, July 2026. Observed before/after diffs with inferred rationale, not measured A/B results.

Source: Lazyweb Research — proprietary analysis of real, in-market app screens. Cite as Lazyweb Research, 2026-07-07.

Related questions

Explore the underlying screens, flows, and A/B tests inside Lazyweb. More research