Introduction:

CareRev is a company that enables nurses to take per diem shifts with hospitals as they appear in the app. In the US, there is a high demand for nurses and inadequate supply.

We lose over 80% of our leads since we ask them to submit a redundant document during the initial sign up process.

Problem:

CareRev invests significantly to get leads into the acquisition funnel, however, the majority of these leads do not end up taking a shift. This is due to several reasons:

·      Speed: Initial sign up can take up to 10 weeks if not longer in busy season. Figure 1 below outlines the sign-up process.

·      Poor UX: After the first step of the sign-up process, a lead must wait several days before they can continue to the next step. This leads to a drop off in the funnel.

·      Inconvenience: The lead is asked to complete a cumbersome process in initial sign up. This isn’t ideal since the same nurse is regularly approached by traditional recruitment agencies and they offer to handhold the nurse through the process. Why use us?

·      Low ROI: Nurses coming in for regions we already have a large pool in.

·      Insufficient staff: CareRev does not have enough nurses to interview all the nurse leads that are interested. This leads to long wait times till an interview can be scheduled.

·      Low Trust: A license number can be the same as the SSN. Nurses are reluctant to share it with an unfamiliar company and this is a high barrier.

My role was to address the problem of leads not taking up shifts by addressing the above causes.

Motivation:

CareRev was expanding rapidly but was unable to fill shifts with nurses despite the large inflow of leads.

Specifically, there were two kinds of expansion that CareRev was undergoing. Firstly, we were attracting new clients in regions we already serviced. In this case we had a large pool of nurses and could easily fill shifts. However, we needed new users, but this was not critical yet.  Secondly, we were expanding into new regions where we did not have an existing pool of users to fall back on to pick up shifts. It was critical for us to get more nurses for these new regions. We did not triage incoming leads for reasons described in the previous section.

In comparison, our competitors would just call the nurses and handhold them through the lengthy process. CareRev needed to get nurses signed up faster into the system to demonstrate value for both users and clients. In our case, that is the ability to pick up a shift to earn money. In fact, this is the single biggest motivator to work with a service like CareRev (based on my findings from a survey I did with marketing with 10,000 ER nurses in a region we weren’t in at the time).

Upon auditing the entire process from sign up to a shift completed, I learnt from the data (using Amplitude) that CareRev lost 80% of leads in the resume submission stage. It became clear to me that we needed to innovate our sign-up process to expand and speed up the appointment of nurses for our clients. 

Implementation and Outcomes:

II kicked off a series of conversations with stakeholders across the company to understand the rationale for seeking a resume during sign up. My designer and I also conducted user interviews among leads that ghosted us after initial sign up and leads that signed up successfully. We did this to get feedback on what the leads thought about the initial sign-up process. Speed was a concern that was brought up by the ghosted leads.

Several iterative changes were required to the sign-up process to address the problems outlined earlier. Below are the outcomes I achieved because of the changes made to the sign-up process.

Removing resume from sign up:

This document was part of the process as it was assumed by the clinical operations team to be part of the Joint Commission Accreditation requirement. After conversations with legal, and understanding the requirement, I discovered that we only needed a “work experience document” not a resume. We could obtain this by asking 3-4 simple questions during the interview, should the lead progress past this point.

After getting buy in from all relevant stakeholders (clinical operations, interview team, legal and others), I updated our interview team’s workflow on the system to include a summary section that draws on the interviewee’s responses to the three questions. Once the interview is completed, a pdf “work experience document” is automatically created and attached to the lead. 

License verification automation:

As outlined earlier, a high intent lead is unable to schedule an interview and must wait several days before they can get to the next step. This leads to a drop off in the funnel.

I negotiated with Nursys (a license verification entity) to get API access to their license verification service for free. We used this to verify the license on the spot during the interview itself.
Looking at the data, over 98% of leads that commenced the process had a valid license. We recognized that we didn’t need to verify the license in Step 1. We only needed to verify the lead if they were a good candidate (passed interview) and wanted to work in a region where we needed more users.

As you can see in Figure 2, the average time for license verification dropped significantly shortly after the automated license verification was released in phases across all professions.

Note: for sake of brevity, I have not discussed the new user type and distinction that had to be created (professional vs lead vs professional lead) to allow for this update.

Triage leads in priority regions:

We added the ability in the system for the operations staff to select which US states and professions/specialties are high priority areas for the business along with existing business (low priority) or not recruiting.

In addition, we updated the sign-up page to include a question asking the lead which US state they would like to work in. This is shown in Figure 3 below.  

As a result of this change, ~95% of interview slots went to priority regions, ~4.5% for existing regions, ~.5% for not recruiting (had scheduled interviews earlier).

Account creation in sign up:

By adding account creation in sign up and preventing users from signing up multiple times, we were able to also check if a phone number was registered and create a forgot password loop. We reduced duplicates by 80% which helped open the interview funnel. This is shown in Figure 4 below.

 The leads had a much better user experience, and we have a greater number of leads being converted to users that can take shifts.

These cumulative changes along with other teams’ work helped CareRev increase gross revenue from $100M to $250M. Just the license verification and triage features mentioned increased LTV by $16M+ per year. Prior to this ~27% of interview slots were wasted on leads who couldn’t pick up shifts.

Here is another Case Study on how I launched a new business line and going from $0 to $50M in annual revenue