Case Study: Pending Requests Workflow

Overview
When a teacher registers with Yoga Alliance, they need to submit a request to their school in order to confirm their yoga teacher training certificate. However, if the school does not review their request within two weeks, it is sent to the Yoga Alliance credentialing team for review. This queue is only reviewed once a month and contains 300-400 requests on average.

In addition to the volume of requests, the credentialing associates responsible for managing the queue were struggling to find the information they needed to confirm the requests, including historical data for that school and other requests submitted during that time period.

Goal
Store all training certificates for a school in a single table and improve the pending requests workflow in order to reduce the amount of time credentialing associates spend reviewing these requests.

Discovery
I sat down with one of our credentialing associates to understand the process she followed when reviewing certificates. I timed her completing two scenarios. First, she reviewed a certificate where the trainer signature matched the name of a school staff member or trainer listed on the account (scenario A). Next, she reviewed a certificate where she could not match a trainer name to the signature and needed to compare it to other certificates from that school (scenario B).

Scenario A took 5 minutes and required multiple instances of manual copy and pasting. Scenario B took an additional 10 minutes (15 minutes total) and required a lot of guesswork. Teachers often register with Yoga Alliance years after completing their teacher training, so review dates can be years after their completion dates. It could take an associate 4-5 searches to find another teacher who completed a training at the same school around the same time.

Design
My first proposed workflow included updates to existing reports and tables, as well as adding a new certificates table to store all school certificates. The main improvements in the first iteration were:

• New table design to improve readability of pending requests queue and school staff report
• Email in pending requests queue links directly to user's account
• Stored pending requests in new table with certificate viewer on user's account so the certificate doesn't require download to view
• Consolidated school staff role and trainer history into a single report, accessible from a button on the manage school account page
• Added school certificates table containing certificates for all confirmed trainees of a school, in addition to any sample certificates the school added
• School certificates table is searchable by date, in addition to who confirmed that certificate

I mocked up a simple InVision prototype for the associate to click through and provide feedback. She confirmed that the proposed updates addressed her biggest pain points, and after some minor language changes, I sent off the wireframes for approval.

Because this was an internal project, we wanted to keep development costs low. While the school certificates table was approved in its entirety, I scaled down the proposed UI changes to existing pages and reports to focus on the most time-consuming issues. The following changes were made:

• Updated column headings in pending requests queue from database headings to language more easily understandable to credentialing associates
• Instead of linking the email addresses in the pending request queue, added links to user accounts that open in new tab
• Added trainer history to the school staff report

Delivery
After the wireframes were approved, I managed the workflow between our front end developer and back end development team and tested the work for quality assurance. The updates were deployed in fall 2017.

These changes reduced the total time to review a request from 5 minutes to 3 minutes for scenario A (can validate signature), and from 15 minutes to 6 minutes for scenario B (can't validate signature). If the full workflow changes were implemented, I estimate another minute would be reduced from the total time in each scenario.