Problem Statement
O.C. Tanner provides tools to their clients that enable employees to give peer-to-peer recognition in a variety of ways. The primary method is to use the legacy platform, Victories, and the new platform, Culture Cloud. Both products currently share a method of giving.
We needed a clearer picture of how users progressed our “Give Flow”—the process of starting, selecting, and sending recognition to a coworker. We also needed to identify critical junctures in the flow where we had a higher rate of drop-offs, so we could dig in and see if there were some “quick fixes” we could implement.
Users and Audience
This was a project where the primary stakeholders were me and the other members of the product triad I was a part of and our leadership team. The key audience was our customers’ employees.
Roles and Responsibilities
Worked with my product triad partners, a product manager, and an engineering lead for the team. My product manager and I did much of the discovery work, from reaching out to our data team and making requests, and analyzing the data for certain scenarios to researching and diagramming the flow.
Scope and Constraints
We weren’t sure what the best approach was going to be, so we had a few false starts. From there we tried to identify elements we felt we needed to know more about. As a result, we generated several artifacts that didn’t prove useful given the early stages we were in measuring user engagement and progress in our application.
Another challenge we faced was getting accurate data. There is a fork in the flow at the beginning, so finding parity in the 2 flows to compare usage was challenging. We also needed to create an aggregated view of our data outside of our analytics solution since we couldn’t have side-by-side funnel comparisons.
Process
The element that started us down a successful path was my researching and diagramming of the existing flow and interactions that we could tag in our analytics solution, Pendo. With the help of our UXR team, I tagged and built the initial funnels we needed to measure a user’s progress through the Give Flow. We also used Fullstory to observe user sessions for additional insights.
Once everything was instrumented and funnels built, we did an analysis of a user’s journey. One of the key discoveries was the drop-off at the end of the flow. We felt we could improve engagement by making a change and swapping buttons. A simple change, but it had a big impact!
Outcomes and Learnings
While the analysis and design were straightforward, we encountered some delays in getting this implemented and released. We learned that we needed to adjust our delivery process and get small changes like this in and released on a more regular basis.
- A 5% increase in the rate of giving recognition.
- This small change immediately increased the rate of people going back into the flow by 200%!
- We also discovered that those who started a second recognition were 90% more likely to complete it a second time compared to the initial completion rate of about 65-ish%.