Local landmark in Humana's homepage banner: +31% clickthrough
Commentary byLiftArchiveOur analysis, not theirsShowing the same banner couple in front of a local landmark, picked by the visitor's location, raised Humana's banner clickthrough by 31%.
Humana's digital test team ran the test, and it was published by MarketingExperiments after a MarketingSherpa Summit 2016 talk.
Only the background changed, so this is a clean read on geo-personalized imagery.
- Banner clickthrough rose 31%.
- Humana said overall conversion through the funnel also rose, but gave no number.
- Sample size, duration and confidence weren't shared.
The change
Same couple, but the background shows a landmark near the visitor.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The banner showed a couple against a generic background that could have been anywhere.
Variant
The same couple appeared in front of a landmark matched to the visitor's location: the Golden Gate Bridge for the Bay Area and the Statue of Liberty for New York.
Why it likely won
Our reading of the result, not the practitioner's claim.
- A familiar local scene may make the brand feel like it serves the visitor's own area.
- Insurance plans often vary by region, so a local image hints that the content is relevant.
- Recognising a landmark can catch the eye in a banner people usually skim.
Caveats: Sample size, duration and significance weren't shared. The funnel conversion lift was reported but not quantified.
Who ran it
Humana ran this test and published the results (MarketingExperiments). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Humana →
How to run this test on your site
- Hypothesis
- If we show imagery tied to the visitor's region, then banner clickthrough will rise because the offer feels local and relevant.
- Primary metric
- Banner clickthrough rate
- Guardrails
- Funnel conversion (quote or lead submissions)
- Page load time with personalization
- Who should test it
- Lead-gen businesses with regional products or service areas and enough traffic per region. It adds little if your offer is the same everywhere or traffic is concentrated in one place.
Questions
What is geo-personalization?
Showing different content, such as images or offers, based on where the visitor is. It usually relies on IP-based location.
Does geo-personalization need a lot of traffic?
Each regional version needs enough visitors to judge.
Start with your biggest regions and a generic fallback for the rest.