Booking Outdoor Experience Online
Optimizing the online booking process for a startup using behavioral science
OVERVIEW
Venku is the fastest growing marketplace to discover and book top-rated, high-quality outdoor experiences across America. A variety of experiences led by expert hosts can be discovered and booked through the site.
PROBLEM
Venku found that users rarely book on the site without first speaking to a member of the sales team or the host. We worked with Venku to determine what exactly the customer needs to feel confident when booking and capable of completing the booking process on their own so that they can continue to build and scale a marketplace for outdoor experiences. Our research questions included:
Who are Venku’s customers and what mental models do they hold of online booking?
At what point(s) in the customer’s journey does she decide between getting assistance or booking independently?
What behaviorally informed strategies can we employ to reduce friction and increase trust?
RESEARCH QUESTION
How might we improve the search process so that users can confidently find an ideal experience, as measured by booking conversion rates and contact host rates?
Client: Venku
Duration: January - May 2022 | 4 months
Location: USA
Role: Behavioral scientist, client liaison
Team: 4 behavioral scientists
Skills: mixed-methods research, competitor analysis, affinity mapping, prototyping, synthesis, usability testing
Tools: Miro, Figma
PROCESS
DISCOVER
Aimed at building a deep contextual understanding, we conducted mixed-methods research.
We reviewed literature from behavioral science, drivers of trust online, and emotional design to give us a better grasp on the factors that influence people’s online choices and inform data collection.
We analyzed demographics, website stats, and a Competitor Analysis to understand the Venku customer, website features, and the strengths and weaknesses of the booking process.
STAKEHOLDER INTERVIEWS
We conducted semi-structured in-depth interviews with Venku to learn who the current customer is; frequent questions they receive; and their vision for a better user experience. We also conducted customer interviews, including a contextual inquiry exercise to observe one user book a hunting trip and document their process, problems, and any confusion or other emotions during the activity. We wanted to understand why customers were unable to book trips without the help of the sales team and this allowed us to observe the exact moments where customers ran into issues.
Thoughts from users
“I type ‘Montana’ in the keyword box to filter results but other states show up.”
“The details of this hunting trip are unclear — will it only be my group or combined with other groups?”
“The sales team can hear what I’m interested in and immediately curate recommendations, it takes too long to do it myself.”
USABILITY TESTS
We ran usability tests with 15 participants who had never used Venku. Each participant received one of three scenario-based tasks they had to complete on the website. We used the think-aloud method to gather feedback on their experience looking for a hunting or fishing trip, ideas for improvement, and reasons why they might contact customer service. The main takeaways included:
Lack of consistency across listings as well as lack of transparency regarding the payment policy decreased user trust
Pictures were unhelpful for determining what to expect on a trip
Improved map functionality and enhanced filtering would make finding the experience much easier
QUANTITATIVE RESEARCH
Next, we designed a survey to quantify user satisfaction with the functionality of the booking process and what features they’d like to see. Our sample (n = 23) included new customers and customers who had booked at least 1 experience on Venku in the last year.
Results
Users skew older and were used to booking through traditional methods (e.g., at trade shows)
Customers primarily use the keyword function to filter by the specific species they want to hunt/fish
Only about 50% are satisfied with the search process
The top requested tools they’d like to see included customer reviews, improved ability to filter, improved listing features, and a more personalized feeling.
DEFINE
AFFINITY MAPPING
To make sense of the data gathered in our discovery phase, we used affinity mapping. Examples of themes included the need for a human element (which Venku delivered on), frustrations with the user experience, and mistrust of the hosts and experiences. These findings helped us determine the main barriers and facilitators to booking online and informed our recommendations.
BEHAVIORAL MAP
I created a behavioral map outlining the journey a typical user faces as they book an experience on Venku’s existing website, including the barriers and opportunities we identified in our research. We identified three main sources from which problems stemmed: host inefficiencies, search functionality, and listing inefficiencies.
HOW MIGHT WE…
Venku’s customers were older (35+), so we hypothesized that building trust with these users would require the website feel more hands-on, offer streamlined but personalized features, and include signals of trust more prominently. After identifying three major pain points discouraging users from booking an outdoor experience online, feedback from our client helped us focus on improving the search functionality. We wanted to know how might we:
Improve the search process so that users can confidently navigate the website to find an ideal experience?
Make searching easier instead of overwhelming and unclear?
Recreate the human element that talking to the sales team offers users?
DESIGN
DESIGN GOALS
Our approach followed the Behavioral Insight Team’s EAST framework. If you want to encourage a behavior, make it:
ITERATIONS
After narrowing in on search functionality, we jumped into ideating website improvements. We ran an ideation session with 15 class members to get our ideas on paper and create lo-fidelity prototypes. My team created and presented 16 mid-fidelity prototypes to our client, after which we guided her through a prioritization exercise to narrow these down to 7 which would become part of our final recommendations.
Lo-fidelity prototype exercise
2x2 prioritization grid
DELIVER
Take a look at some of our solutions on Venku’s home page:
