Quizless Funnel Analysis
Techstyle: Fabletics // Mar. 2020
After the success of the checkout redesign, I was asked to analyze the sign up flow, identify potential optimizations, and present the findings to the executive team.
The initial proposal was a powerpoint deck, however rather than just upload images of the deck and for the sake of a better experience, I’ve converted the deck into the following project page.
Quick Summary
The Problem
How might we improve this experience to create a more efficient sign up flow while increasing sign ups?
The Question
Every Techstyle brand forced users to take a style quiz upon sign up. Internally it was beloved, but our users were dropping off the flow before they even got to a shoppable grid.
The site in it’s current state depended on gathering certain info from the quiz, and was built in a way that an unregistered users couldn’t browse or shop in a logged out state.
The Hypothesis
If we allowed users to shop our site without registering (including adding to bag/viewing cart), we will increase conversion due to better demonstrating our value proposition on site.
Identifying Our Focus
Experiences need a focus to be effective. While casting a wide net and trying to “get everyone” sounds good, it more often leads to making everyone unhappy. So what were we as a company trying to gain from the quiz?
Starting at the Very Beginning
Empathizing with our users is just half the battle. Getting stakeholders to empathize, and then make changes is even more difficult. This required providing common feedback and reasons users did not like creating accounts
Why don’t users like to register?
They may not be planning on returning with intent of just making a one time purchase
Maybe dislike reg. In general and not want more spam/another account to remember
Users are trepidatious about how their info gets used/saved
Introduces room for user errors (filling out form wrong and abandoning, etc)
Unsure of deletion process
Too much work
Not understanding why certain info is required
Interaction Cost
The interaction cost is the sum of efforts — mental and physical — that the users must deploy in interacting with a site in order to reach their goals. Things like:
Reading
Scrolling
Comprehending info
Clicking
Typing
Page loads/wait times
Attention switches (aka from simple quiz to full reg)
Giving up personal info
Friction and the Sign Up Flow
Once we had aligned that reducing friction was the area of focus for this analysis, I got to work. The first steps were to properly define what fricton looked like in our sign up flow.
This consisted primarily of:
Steps to completion:
Number of steps or series of pages that a user is expected to pass through
Information Cost:
Number of fields that a user has to fill up
Effort Invested:
The number of decisions the user must make, and the number of additional activities required
Control Sign Up Flow
I worked with our Google Analytics team to better understand where and how the majority of users were making it to our site and put together a flow chart to get a clearer understanding.
Breaking it Down
Purpose driven friction is friction that either benefits the business or the user. Regardless of it’s value, it is friction nonetheless, but defining it’s purpose helps identify if it’s worth the trade off or not.
Below are all the points of friction a user had to get through just to reach a shoppable grid:
Steps to completion
Gender select pop up
Get started cta
Quiz steps (5)
Custom size info (6)
Pop up
Final info (3)
Information Cost
Personal information of user
Personal preferences (color, workout, etc)
Bottom, top, bra sizes
Zip code
How they heard about us
Name
Email
Password
Effort Involved
17 required friction/decision points (15 in quiz)
8 we currently use
Bottom size, top size, bra size
Zip code
“How did you hear about us popup
Email, First name, Last Name, Password
Green = Purpose Driven
Competitor Analysis
To paint a clearer picture I compared our flow to direct and indirect competitors. I was mostly interested in if registration was required for checkout, where it was in flow, and what information was needed. These were my findings:
It was immediately clear we were outliers in our industry.
With Registration Barrier
Fabletics
Target
Without Registration Barrier
Adidas
Outdoor Voices
Lululemon
Gymshark
Athleta
Carbon38
Old Navy
Nike
New lead sign up location across competitors
Takeaways
By moving registration to checkout we provide the users opportunity to shop a frictionless experience and find the value in creating an account in the right context
Of competitors who have a “sign up” callout in a new lead flow, 7 out of 8 display it when a user is in checkout
Recommended Updates to Flow
In doing this analysis I discovered that less than half the friction points were purpose driven. The majority (questions like “What type of workouts do you like” or “preferred colors”, questions that should have been returned back to the user in way of personalization, were not being used or even collected by the business.
The he recommended updates basically “cut the fat” (the useless pieces that were just extra hurdles) and allowed users to enter discovery mode immediately.
Control
Recommendation
Part of this updated flow included moving the quiz from the fore front and making it optional on the shopping grid.
One interesting finding was that the majority of users who saw the control flow stated they’d abandon upon seeing the quiz, however when a new group was tested the majority of users organically noticed and mentioned one of their first actions would be taking the quiz. The freedom of choice was seemingly the missing element.
Other improvements included collecting info (that used to be collected in the quiz, like size) elsewhere on the site; Caching size inputs on PDPs was much more natural and the user experience 0 friction since size input is an understood requirement while shopping.
In a similar fashion, all users understood that in order to receive an online purchase, their address/credit card/etc was required (or “purpose-driven friction”). We’d have much better luck having users create an account when they’re already inputting the information, presumably because they’ve found items they want, rather than making them create an account before they’ve even seen the sight.