Feature Onboarding
A/B Testing
tl;dr
Problem
Many new hosts were signing in, selecting the goal “Find Cleaners,” but they ended never creating a cleaner search, leaving the platform before taking their first important action
Goal
Guide users through their first cleaner search experience, increasing the number of cleaner searches created and reducing early churn.
Impact
Increased cleaner search creation by almost 9%
Deleted accounts without creating a search dropped by 5%.
Established a scalable component to onboard user on features they've never seen before
Context
Churn. That was all about during Q3.
A lot of new hosts were leaving the platform before completing meaningful actions, becoming one of the company’s top concerns
For this project I was assigned to work on reducing churn on the early stage on this specific flow:
Hosts who signed up through the “Regular Sign In” flow
Selected the goal “Find Cleaners.”
But never created a cleaner search.
Hosts were leaving before taking their first important action, without a chance to see what the product could offer to them.
Framing the problem
Discovering the issues
Research Team: We checked the comments from deleted accounts surveys, we saw a lot of words being constantly repeating such as “Not clear information”, “Not what I expected” or “Don’t know what to do”
Data Team: Get to know on a Mixpanel funnel to see how many users who signed up, selected “Find Cleaners” as their goal and created a search:
Putting on the user’s shoes
I also went through the flow from the user’s perspective to see what they experienced after the onboarding. What we found:
After selecting their goal, users had the option to explore the product on their own, which took them to the Dashboard. This created friction, because it wasn’t the place to start their goal
We noticed there wasn’t a clear path to the page users needed to reach their goal.
While creating a search, we were introducing concepts that weren't really familiar for them yet
What we assumed
It became clearer that this wasn’t about users not being interested, but about a lack of clarity on how to continue:
Users didn't know how to continue
The experience was a little overwhelming because we were taking them to a place that is not related to what they were looking for
From Insights to Action
Defining an hypothesis
We believed that by clarifying how the cleaner search works, removing friction, and providing a more straightforward experience after the onboarding, users would know what to expect and be more likely to create a cleaner search, which was their primary goal:
Host confusions and Insights
To approach that hypothesis, I started to think ways in which we could add value to users, with questions like:
What would they want to know before creating a search?
What sort of things would them stop from creating a search?
To find some insights, I partnered with the Sales team. They had hundreds of calls with hosts and they helped us identify their concerns, common questions and bias. We organized their feedback into 4 specific topics:
The Changes
To make things easier for hosts, we introduced two key improvements to their flow:
A direct path to the Cleaner Search
After completing the onboarding, users who selected the “Find Cleaners” goal would be now taken directly to the Cleaner Search page, skipping unnecessary distractions
The Feature Onboarding
Instead of landing directly on the first step of the Cleaner Search wizard, users would now see an introductory page, explaining what the Cleaner Search is and what to expect from it
It wasn’t a tutorial, but a quick moment to reduce friction and provide clarity so users can complete their first meaningful action
Results & Impact
Business Impact
Before Feature Onboarding
28.36%
Searches created
After Feature Onboarding
37.29%
Searches created
Although this didn’t guarantee that users would complete the full conversion flow (they would need to accept a cleaner and then complete a project), it made a big difference because fewer users were dropping off at the very beginning, and more were entering the funnel of possible conversion
Design Impact
The Feature Onboarding became a reusable template we could use across other parts of the product, helping explain other key features
This would help not only on those users whose main goal was to find cleaners, but for those focused on managing cleaners through other features such as Checklists, Property Problems, and more
Final Reflections
Design can be the last phase of a designer
It’s better to gather insights first, as they help define the real user problems. Then we can start shaping the right solution.
Simplicity matters
One single screen and a tiny logic tweak can make a huge difference!
Other Projects
Quality Center
A dashboard to help hosts improve their Airbnb cleanliness rating, driving 3× more monthly integrations
Recommended Cleaner
New cards using data insights to help hosts trust and choose the right cleaner, increasing conversions by 5%