Install to Registration Conversion
Client: Ula | 2022 | UX
The previous self-onboarding flow required users to navigate through 10 steps, resulting in significant drop-offs at certain points according to data. Our goal was to enhance overall conversion by identifying the issues and implementing solutions.
Initial Brief
Increase the install-to-registration conversion % of self-onboarding users. Identify the friction points to minimize drop-offs.
Looking at numbers
With our data analysts' assistance, we noticed that in the whole process, we were losing a lot of people at three key points: when they had to enter their name, mobile number, and store details.
To get a better grasp of these drop-offs, our research team in Indonesia hit the ground and spoke with few store owners. They asked questions related to the onboarding process and the overall steps involved in using the app, aiming to understand their thoughts and feelings about it.
Findings
Our research yielded clear results. Store owners have established trust with their longtime local wholesalers. People generally consider using the app for two primary reasons:
First, to explore the variety of products Ula offers, and second, to check if Ula's pricing is more competitive than what they get from traditional suppliers.
This sheds light on why we observe drop-offs at points where customers are hesitant to share personal information like their name, phone number, or store details just to know pricing and product availability.
Pitching our approach to Stakeholders
As a business, Ula aimed to gather user data like mobile numbers and names for database purposes and user identification, while store details and location were essential for displaying available products in a specific area and facilitating deliveries.
However, recognizing that customers might be hesitant to share these details upfront, our product manager and other stakeholders reached a consensus that requesting a precise address during initial registration might not be practical.
Consequently, we made the decision to streamline the registration process, only asking for essential information: the phone number and location access to display products and prices in the user's area. Additional details and address input were moved to the checkout process, where users are more likely to complete these steps when their motivation is stronger.
Moreover, we introduced a guest flow option at the phone number input screen, giving users the choice to explore the app without sharing any personal information, respecting their need for anonymity.
New User flow
Given our team's dispersed global presence, this project, like others, required a collaborative approach. However, this project primarily involved a deductive design process, focusing on trimming features rather than introducing new screens. We promptly addressed user flow issues and implemented necessary adjustments, with only minor UI changes made along the way.
Visual Design
Say hello to the new onboarding flow, this was our final UI that went live. Adding a prototype here to understand the flow better.
Additional Tweaks
In the process, we also took care of a few minor issues in the flow.
For example, we streamlined the registration and login flows. Previously, if a user's number was already in the database and they tried to register, they'd encounter an error message. Now, we've merged these into a single flow where the system recognizes the number and automatically guides the user to the login or registration process as needed.
We also made some user-friendly changes like defaulting to WhatsApp for OTP verification since almost all our users have it. The developers made minor API tweaks to ensure OTP delivery consistency.
We simplified the process further by eliminating the need for users to specify their store type since it didn't provide significant value to Ula and caused some confusion for our customers as to what category they should put their store in.
Additionally, we added autofill for addresses using location access.
Addressing user feedback, we updated the Play Store images with the support of our marketing team to provide a clearer explanation of the Ula app's features, especially since it was relatively new and two years old, some users complained that they didn't clearly understood what Ula does.
Following up
We introduced the new flow incrementally, starting with a pilot in a few provinces before gradually extending it to all regions. This phased approach resulted in substantial enhancements in registration completion rates, as illustrated below.
In this project, the research proved to be invaluable in pinpointing areas for improvement that didn't necessitate additional engineering work. These insights might have remained unnoticed had we solely relied on quantitative data analysis.
