The Knot - CATEGORIES
Product Design, User Research, UX, Visual Design
I was leading product design for the consumer facing side of The Knot's local marketplace, where we connect couples to wedding vendors. Our biggest challenges were innovating in a product area that produced the most revenue for the company and juggling the needs of a two-sided marketplace.
At the beginning of 2017, I led a zone wide experience mapping session. The activity of mapping builds knowledge and consensus across teams and stakeholders, and the map as artifact allows you to create and support better user/customer experiences.
Intention of this exercise:
Build empathy and understanding in actual user/customer needs
Identify specific areas of opportunity to drive ideation and innovation.
Provide a common “big picture”
Help break down silos with cross-squad collaboration
Help bring focus
Reveal unnoticed value-creation opportunities
Prep & Planning Timeline: 4-6 weeks
Duration of Mapping Session: 3 days, allot 2-3 hours per day
Organizing Team: 5 people
Participants: 4-12 people, cross-disciplinary (design, research, product, engineering, marketing, etc)
From Insight to Action
From this zone wide experience mapping session, our squad identified key areas of opportunities within the product area that we owned.
Challenge: Our product area falls within the "research" and "validation" stages. During these stages, we found that our users are often paralyzed by choice, frustration, and stress due to information overload.
We know that when users apply filters they convert to quality leads at a higher rate, but only 17% of users interact with our existing side bar filters.
Performance of the category pages is poor, which negatively impacts the user experience, especially on mobile.
In addition to higher bounce rates on mobiles, the poor performance likely translates to erosion of SERP rank and organic traffic.
After a cross-functional Design Studio, our team converged on these ideas in order to solve the challenges.
Help users move down the funnel faster and feel less overwhelmed
Test horizontal filters and various facets to help users find what they're looking for faster
Experiment with keyword search or an "ad-lib" style prompt
Embed cards that ask users for information to personalize their result set.
Support a better browse experience
Utilize an image carousel in vendor cards to allow users to "quick preview" multiple photos in a card to get an idea if a venue fits their style.
Allow users to browse venues via a map view
Acquire users from organic search
Move vendor bookmarks to the marketplace and improve the "Favorites" experience. Value prop would be to create an account to save your list to share with your significant others or parents.
PROTOTYPING AND TESTING
Unmoderated user testing (usertesting.com) with 8 participants utilizing both Android and iOS devices. A majority of users indicated that they used Google as a primary resource during their venue research efforts. 2 users had heard of The Knot and 1 was using Wedding Spot.
Do updated filters help couples find the vendors they’re looking for more easily?
To gain insight into which mechanisms resonate the most with the user’s current mental model: guided keyword search, facet by style, guided cards.
Validate if users find a map view valuable during this part of their journey.
Guided cards only received one or two interactions but were used in instances where users had higher intent. These users were later in their venue research efforts, so they were apt to using filters and other functionality right away to narrow down their results based on criteria like pricing, capacity, and amenities
Users that were much earlier in their research efforts were apt to spend more time browsing results based on photos. There is an opportunity explore how couple education and wedding style may fit within this browse experience.
Before going into the user test, users seemed to have a general idea of what their style was but we found this be fluid during browse, so top facets are not the best UI treatment to help users explore or discovery this. 1 user appreciated that we were able to "articulate" different styles that she could not do herself.
Only 2/8 users navigated to the map view during their search. Despite utilizing a floating material button, most users were blind to this. Our team did not have deep hypotheses around this idea before testing, but these early results continue to validate that when users are in browse mode, they prefer an image-oriented experience.
We validated that our top filter sets like capacity and price meet users' expectations, though a few noted that price range would be more valuable than our current "$" expression for price. 2 users wanted to filter or search by availability.
There's an opportunity to improve the taxonomy for other filters like "Wedding Categories", in which several users did not understand.
We currently do not support a sort feature on the live site. We found that users wanted to sort by price or distance. No one commented on or tried to sort by "best match".
There is more work to be done to continue to validate keyword search as this is a static prototype. 3 users that navigated here found guided keyword search to be intuitive.
Conversion rate to quality leads
Are we doing a better job matching users with vendors, thus increasing the rate at which couples are contacting those vendors?
Are more users seeing and interacting with the filters?
Click-through to vendor
Are more users getting results that are compelling enough for them to click through and view individual storefronts?
Page load time // Page Speed index (especially mobile)
Are the category pages performing on par with industry standards?
Are we providing a more engaging experience to our mobile users?
Pages per session (mobile)
Bounce rate (mobile)
Avg. session duration (mobile)
I left my role at The Knot before we implemented all phases of this redesign. However, I was involved in the first phase where we refactored our codebase and updated the look and feel of the page to match our new design system. This phase of the redesign improved our page speed score from 60 to 80. NPS score on mobile improved from 43 to 48.