Redesigning an onboarding flow to increase customer success rates

Overview:

A successful onboarding process is critical in gaining customers and as a new financial institution on the market, Volt wanted to increase the first time success rate of customers who were creating accounts. An improvement was proposed by business to add an additional vendor to the onboarding flow that could assist Volt in verifying a customer’s identity and as the UX designer assigned to this problem, I explored the problem further to assess if it was the best option for customer and business.

My roles and responsibilities:

The bulk of this project took place between January to March 2020 with Volt Bank. I worked on uplifting the onboarding flow as the point UX designer.

Problem statement:

How might we increase success in customer verification without elongating the onboarding process, maintaining trust with the customer, and with minimal cost to the company.

 

Users and audience:

This piece of work was targeting new customers of Volt bank. At the time, Volt was in beta phase doing a slow roll out to their waitlist but this would also affect future customers beyond that, specifically the target user of Volt bank: millennials.

Scope and constraints:

As this was an uplift to an existing design, the placement of the flow needed to remain the same. The uplift needed to be within a specific module of onboarding, using primarily existing components from the Volt design library. The new vendor also had specific requirements of use that had to be catered for.

Process.

Investigation:

With a product in the market, I wanted to first understand why customers were failing the existing onboarding process. This would help me identify if there were points in the flow that could be revisited before an additional vendor was required at a cost to the company.

After discussion with my Business Analyst (BA), we chose to approach the Operations team who had been directly troubleshooting customers since launch. With their help, we were able to identify when existing customers were most commonly contacting Volt, what the most common problems were, and how the team were assisting them. Together, we learned that customers were primarily struggling with data entry: be it customers misentering information or built-in constraints (e.g. input restrictions or limits).

With this new information, I took a holistic view of the whole onboarding process and began to mark specific points in the existing flow that were commonly causing customers pain.

This map was used to get a big picture view of what the cost to the company would be if we were to make changes.

In order to make a fair assessment of the best solution for Volt, I also looked into the suggested vendor so we could understand how adding a new partner would increase customer verification success if no (or minimal) changes were made to the existing flow. We learned that after the customer had provided their details, the vendor would run them through another database, potentially increasing the success rate of that customer’s verification provided they had entered in the correct details to begin with. That is, the vendor would not help customers who were failing due to data entry: the existing number one cause.
However, putting data entry aside, we also mapped out the statistics of the customers that this vendor would assist, finding the percentage to be alarmingly low whilst still charging the company for each check, successful or not.

With the resulting investigations, a successful case was presented back to the business to focus Volt’s attention on addressing the known pain points through an uplift of the existing design instead of adding an additional vendor.

The uplift begins:

With a new brief and pain point map ready to go, I began prioritising the customer pain points and began the uplift by rearranging the order of existing screens, consolidating where I could. This was done to better frame the information we needed to collect without requiring more instruction screens and to reduce the cognitive load for customers. On this map, I kept track of what pain points we were addressing with the adjustments, and flagged what still needed consideration.

I then revisited the copy with Content to add more supporting text to each screen wherever we collect customer data to provide the customer with more context and highlight the importance of letter-accurate information.

Out with the old:

Alongside the Engineering department and my BA, the next step was to understand why certain restrictions were built into the screens originally and what could be easily adjusted. We learned that for time, certain decisions had been built that were not actually applicable for every screen but were impacting pass rates. Alongside my BA, we then decided to start from a clean slate and contacted our existing vendors to learn what previous requirements we could lift, learning that we could increase most of the text character limits and expand keyboard inputs beyond alpha-numeric characters (i.e. to include hyphens, apostrophes, and spaces in names). With the lift of these constraints, I then opted to include a dynamic character limit on each text input screen so we could be more transparent to our customer.

These changes allowed the customer to more precisely enter their details as it may appear on their ID and lifted the character restrictions for customers with longer names, which were both impacting the success of customer verification.

Check in:

In addition to all the new information discovered and changes made, I again revisited my pain points map and looked at any point in the flow that could still cause customer frustration.

It was at these points that my BA and I went back to the Operations team to ensure that if a customer got stuck, they could easily contact Volt’s Customer Care team to find support. With this confirmation, I then added a quick access lifeline directly with the Volt Customer Care team at any anticipated pain point during the verification process. 

Prioritisation for build:

Having implemented these changes and feeling confident I had addressed the key concerns, I then worked with the Business Analysts and Product Owners to break down the uplift and ensure that the appropriate changes were implemented in the Engineering scrums according to priority within a timely manner.

Outcomes:

The result of this investigation and subsequent uplift was an updated onboarding verification flow that increased the chances of a customer’s self-service success by 4.65%, involved consistent collaboration and negotiation across departments, did not unnecessarily employ the services of an additional vendor, and most importantly, was at all times customer focused to ensure an improved and supported experience.

Lessons:

This project highlighted the importance of two key areas: the importance of collaboration, and the importance of listening to real users. Without both of these, we wouldn’t have found the real problems and complete this uplift. Investigating the pain points of both customers and Volt employees was critical in improving the design for both sides of the app, and while no design is ever final, we can make sure we are listening to find ways to improve upon experiences.