BedandBreakfast.com responsive search results

Responsive search results

Background

BedandBreakfast.com is the world’s biggest aggregate site for B&B and inns, but online booking (OLB) is not required. Instead, BBcom provides external links to the inns’ websites.

The “Visit website” CTA had and continues to be an ongoing business debate: Is it worth sending travelers offsite? But the link continues as big seller and key metric for innkeepers.

Plan

Goals: Drive better traveler conversion on the search results page; balance the needs for SEO/organic traffic growth; show better tier differentiation to aid marketing; responsive experience

Expected impact: Increased click-through on key metrics (inn website’s visit, property page visit, online bookings, checking availability modal; a decrease or flat bounce rate; SEO performance improvement; an increase of membership tier upgrades

Google also re-configured their search results algorithms, so the SEO team planned to resubmit search pages and city refinements sitemaps.

Users

The BBcom traveler profile is skews to older couples that are driving. The hope is, as the site design becomes more modern, that we attract younger travelers. Not all B&Bs are lace and doilies, after all!

Prototype

The team started with some quick front-end changes, such as moving an ad above the search bar to the right column. There was also an small amount of copy that didn’t serve a purpose, so we removed the number of rooms per inn, and changed “$XX-XX per night” to “from $XX”.

The search results page didn’t have H1 tags, so a “{City} Bed and Breakfasts” was added to the top. Meta descriptions were also updated to feature the number of guest reviews, property photos, and inns in a city.

I re-layed out the page, updated the UI, and found ways to promote higher-paid tier inns. We discussed adding icons for inn features per listing, but I felt the page was already busy and didn’t need additional icons that people tend to ignore.

One of the biggest changes, however, was removing the primary orange CTA button when there was no online booking and drawing attention to the external website link.

Test& iterate

Hypothesis: The above changes should add a small boost to the traveler booking metrics while keeping bounce rate flat.

This design was ripe for numerous and fun tests. We AB tested these changes and aimed to drive 3%+ increase (test vs. control) in at least one of our key metrics.
Two successive rounds of design changes had very positive impacts when AB tested. When adding these iterative releases together, our results were 2-3x our goal of a 3% conversion increase (depending on the specific metric in question).
Here were some interesting test results:

  • “Inn’s Website” vs “Visit Website”: the latter produced 10% increased bookings
  • Changing “$XX-XX per night” to “from $XX: OLB conversion +19%; website click-throughs +5%; bounce rate -1%.
  • Updated buttons and fonts across the site: +3.7% on bookings; +13% on inquiries; +4.6% on click through
  • Updating styling an layout (search bar, new headers): booking conversion +6%; +4% click through conversion; click-through conversion +2%
  • After the page was release into production, the bounce rate improvement was -7%

Deploy

After deployment, there was an immediate jump in SEO traffic as soon as the page was pushed to 100%.

Graph of conversion after 100% release and comparison YoY

Graph of conversion after 100% release and comparison YoY