Overview

Problem

46% of 14000+ support requests from users were for a specific type of issue that consumed a significant amount of the CS team's time to resolve.

I wanted to find out what could be done to address the high volume of support requests.

Outcome

Research efforts initiated a reiteration of the user onboarding experience.

I presented findings and recommended solutions to Senior Leadership team and various stakeholders.

NG - Situation

NurseGrid is a mobile app made for nurses by nurses and a schedule tool to help medical professionals manage their demanding work schedules.

As a Product Support Specialist for NurseGrid, I got to interact and work with our users more than anyone else on the team. Everyday, I read and managed every kind of ticket; bugs, feature requests, questions, feedback...

As soon as I started at NurseGrid, it didn't take very long for me to see how much valuable "free research" our users were providing us on the daily at their own free will.

I was fascinated with the feedback we were getting but I was even more excited about getting to talk about them with the rest of the team. I thought

"I have these awesome insights straight from our users...who do I need to share them with so that we can start making NurseGrid better for them?"

"Lost" Opportunity

"Lost" User Feedback

"Lost" Product Road Map

"Lost" Build & Functionality

What this high-level "problemobile" told me was that there was lots of opportunity for me to use data and research to raise the user's voice when it came to improving NurseGrid. I got to work by establishing a user feedback loop and processes to encourage better cross-functioning collaboration and strengthen the link between user feedback and product road-mapping.

Discovery

I began by implementing a ticket tagging system in Intercom to get a better grasp on the nature of our tickets and volume. I consulted with several online articles and resources (like this one, this one, and this one) about best practices for collecting and organizing user feedback.

Compiling everything I learned, I collaborated with the CS and Product team to determine what kind of user feedback would be most beneficial for us to collect. I then created definition documentation for our internal team as a guide to help ensure consistent tagging.

I then spent an obnoxious amount of time of going back and tagging every single conversation from the last 6 months (April 2020 - October 2020). Although this took an obscene amount of time, I knew it would be crucial if I wanted to be able to present more accurate number values for any conversations trends.

After I finished tagging, I was able to get a better idea of what our users wanted to talk to us about. One trend I noticed in particular was the volume of Account Deactivation requests from our users. Sure enough, I found that we had received a definite increase in the number of users reaching out requesting that we deactivate their accounts.

I had my own personal suspicions that I thought could explain this rise but I couldn't really know without collecting some more data.

I reached out to our Senior VP Scott McQuigg to let him know about my findings and provide some additional insights about the challenges NurseGrid was facing, based on what I was seeing in Intercom. With his enthusiastic support, I started working with the amazing Kristie Focht, NurseGrid's Controller and Data Analyst, to dig deeper into user activity and churn.

Questions

My goal was to determine what the churn rate was for new NurseGrid users and what pain points may exist for those users that are contributing to that churn. In order to succeed, these were the questions I wanted to answer:

How far into their experience with NurseGrid are new users dropping off?
What is the volume of each type of ticket that we receive from our users?
Based off of new user drop-off rates and the nature and volume of tickets, where do users seem to be facing the biggest pain points after activating their account?

Methods

Survey Design: I then designed a TypeForm survey that was passed along with every response to account deactivation requests so that we could start collecting some further feedback on why users were leaving and what features were missing that would encourage users to stay.

Based off of the little feedback I had previously seen in Intercom, the biggest reason that users reported as to why they stopped using NurseGrid and wanted their accounts deactivated was because "they didn't use it enough."

Data Analysis: I wanted to find out when exactly new users were "not using it enough" after activating their accounts to see if there were any patterns, which could then indicate which part of the experience journey may be potential roadblocks.

I asked Kristie to pull the following numbers:

# of account activations by month for the last 6 months.
# of those accounts who then added worksite within the same day of activation.*
# of those accounts who added a shift to their schedule within 1 week from activation.
# of those accounts who added a shift to their schedule within 1 month from activation.
# of those accounts who added a shift to their schedule within 2 months from activation.

* For context, users are required to add a worksite to their account before they can start adding shifts to their schedule.

With those numbers, I put together a monthly funnel for each of the previous 6 months (September 2020 - February 2021) to show us the rates at which users were dropping off within different timelines from their account activation. I also put together some charts of customer support conversation, taken from the Intercom tagging system. This was to see if we could maybe connect what users wrote in to us about the most and where they may be facing the biggest obstacles with NurseGrid and therefore eventually stop using it.

Results

After collecting all this information and connecting some dots, there were some of the insights I discovered:

66% Average drop-off rate of within first two months for free users.
Worksite Requests are overwhelmingly the highest volume of tickets from users.
Worksite Request process may be an obstacle in the onboarding experience for new users.

Presentation: When I showed Scott my findings, he found them impressive enough that he requested I present them to the Senior Leadership team and other various stakeholders. I put together a slide deck which has now been used to onboard the new Director of Product and has now initiated a reiteration of the new user onboarding experience.

Impact

This project was a great opportunity for me to experience first-hand the importance of looking to your users for answers and guidance. It taught me how much impact is possible if you are resiliently curious, you ask the right questions, and you learn to speak the language of your various stakeholders.

To learn more, let's connect!