User-Driven Onboarding

The Zen Educate onboarding process is designed to help prepare and place educators in their dream supply roles but this can be lengthy and requires manual effort from the user, resulting in a low completion rate for users. In this project, we embarked on the journey of optimising the onboarding experience for users and the business.
Client
Zen Educate
Category
UX Design, User Research

Zen Educate is an online platform that helps connect teachers and teaching assistants with their dream supply work. However, before teachers can start claiming jobs, we require users to upload all necessary documents to ensure we’re placing qualified candidates and protecting the safety of our schools and children. While this process may be lengthy and require some manual effort from the user, it’s a necessary step in ensuring that educators are properly vetted and matched with the right job opportunities.

This onboarding process is lengthy and requires manual effort from the user to upload all their necessary documents in order to get them out to work. Many of these steps are required by the Department for Education for all UK teachers.

‍

Goals

‍

Let's Get Started...
What problem are we facing?

Where We’re Starting...

Here’s where we were starting off with the onboarding process:

Looking Internally

For a user to get verified and start working a job, we require our internal operations team to review their documents and conduct a qualifying interview.

After organising an internal workshop between the customer care team, operations, and product- we identified 2 main pain points:

  • Difficulties in collecting necessary documents; Both product and operations have problems with motivating our users to upload all necessary documents, which are required to get them out to work.
  • Unable to assess user engagement; Our internal team has trouble identifying which user’s are engaged, and therefore ready to start working with us, above the rest.

‍

User Interviews

First impressions matter, and unfortunately user’s weren’t getting the best impression from Zen Educate upon first sign up.

We conducted a 9 user interviews and 1 moderated usability study aimed at uncovering current pain points and prioritising next steps based on users needs.

Gathering User Behaviour & Feedback

During our foundational research, we wanted to uncover:

  1. How Zen Educate’s vetting process compares to our competitors.
  2. What the most important tasks users need to perform are when using the app
  3. Any challenges going through the sign-up process that affected user motivation to continue working with Zen.

Check out the full bank of interview questions.

‍

The Results

  • All participants expressed feeling overwhelmed by the amount of manual effort required to get working with Zen.

"[My experience going through sign-up] was lengthy and it involved a lot of back-and-forth (…) it was more lengthy than I’d had to go through even for more long-term teaching jobs."

"There were elements [of the sign-up process] that seemed maybe a tad excessive"

  • Participants explained they felt a lack of progress and motivation from the platform alone.

"I expected to have a more personal approach. Just some acknowledgment. (...) Once I was contacted [by the operations team], I had more motivation."

"I found it a little bit strange(...)I had to start putting personal things, and there was no feedback. I started filling in my my profile, and there was no feedback in terms of [messaging]. After a while someone called me, and it felt like there was more progress."

  • All users stated their biggest motivator to upload their documents was being promised work over the phone.

"I felt like I didn't really know anything about what was going on because I hadn't done supply before. But then I had a call with [Zen resourcer name] who promised me a job if I uploaded all my documents (…) so that’s when I really ran with it."

  • Candidates expressed dissatisfaction with the amount of information displayed on job cards whilst they’re going through vetting.

"I didn’t even think those jobs you were showing me were real. I mean, they’re lacking so much information I barely look at them."

‍Note: Due to business concerns, we’re unable to display sensitive information about a school until a candidate is verified.

‍

Defining the problem...
What do our users need?

So, Where Do We Start?

We conducted a brainstorming session to plot out pain points, needs, and opportunities, based on the feedback we gathered both internally and from our users.

Our Hypotheses

After this task, we concluded two places to start which address both business and user needs:

Improving information on job postings

Our hypothesis is that by adding more information to job postings, users will be more likely to fall in love with the jobs we're offering, leading to increased interaction with those jobs. This, in turn, will allow our operations team to better gauge user engagement and offer users more suited jobs going forward. Ultimately, this will lead to increased motivation and engagement from users throughout the vetting process.

Work-driven vetting

Our hypothesis is that, by tying the vetting steps to a specific job, users will have a clear understanding of the value they stand to gain by completing the vetting process. This will improve the percentage of user uploaded vetting steps, allowing our operations team to more quickly identify qualified and engaged candidates and get them out to work. Ultimately, this will lead to increased user satisfaction and a more efficient and effective onboarding process for our users.

‍

Building the designs...
Bringing our ideas to life

Designing Our Solution

I started by sketching out the ideal flow we wanted to test out. From our key hypotheses, we wanted to highlight work at an earlier stage of the process to increase our users motivation to complete vetting steps to get out to work.

High-Fidelity Designs
Testing our hypothesis...
Are we moving in the right direction?

About the test

Rolling out this design contains some risk and high development effort, therefore we decided to conduct an unmoderated prototype test on Maze.

‍

Key Goals
  • Uncover current ease-of navigation for candidates navigating to work tab.
  • Observe time-on-task to complete vetting steps when tied to a job.
  • Measure success rate in completing all vetting steps when tied to a job.

3 rounds of testing and 600 participants later, we uncovered some key insights to move forward with implementation.

Outcome

43% uplift in users navigating to, and interacting with a job

By our third round of testing, adding more information to the job postings resulted in a 43% increase in participants interacting with the job.

"The best thing about your organisation is that an applicant can easily decide what job to choose because the way you explain the roles and responsibilities is really clear."

Time-on-task to complete vetting steps slashed in half

According to the usability study conducted, when the vetting steps were tied to a job, the time-on-task to complete them was cut in half for each step. This resulted in 61% of participants stating that they would upload all their necessary documents in one session.

61% of participants said they would upload all their documents in one session.

When asked how quickly the participant would upload each required vetting step (right away, might take some time, wouldn’t complete), an overwhelming majority stated they would upload everything right away. - This is including DBS upload which currently stands at 8% on the platform!

Generally positive feedback from participants

"Everything is laid out so simply. You can see exactly what needs to be filled in and I like how it tells you when it is completed."

"Everything was good. It informs me of what I would be expected of the job, the procedures that I would be adopting when accepting a job from your company and what it is required of me."

"It was quite simple and I enjoyed it more than what you have at the moment."

‍

In conclusion...
What we achieved, challenges, and next steps

Developer Handoff

After receiving strong evidence to prove our hypothesis is correct, we broke down our successful design into smaller incremental steps to reduce implementation effort and reduce risk. The MVP built adopted the idea of populating our job descriptions with more data and prompting users to upload their DBS when first interacting with a role.

Monitoring the results

After rolling out the MVP of our designs, we monitored it for several weeks. As a result, we saw an uplift from 8% to 13% of candidates self-uploading their DBS, smashing our KPI. Additionally, we saw an uplift of 38% of candidates interacting with our more complex job descriptions.

We’re still in the process of incrementally working our way towards the successful designs from the prototypes. With each new implementation, we’re seeing small improvements in our metrics- leading us to believe we’re working in the right direction.

‍

‍

More work
Building A Design System: Aligning Design And Engineering
Read case study
Accessibility-Driven Design Challenge: Redefining Communication in Healthcare
Read case study
Optimising User Experience and Searchability
Read case study