When we were trying to sell a CRM solution to home health care companies, we learned about the challenges faced by home care industry. Getting customers, managing scheduling, hiring caregivers, supporting customers, just to name a few. Amongst these challenges, owners and managers unequivocally said that hiring caregivers is the most time consuming task. Hiring caregivers is particular tricky for home care companies because:
- Transient caregiver workforce. The industry's average caregiver turnover rate is 65% each year.
- Getting new clients. With each new client, the company needs to find a caregiver that is available to work at a particular time and location.
- Client dissatisfaction. The company needs to send in a new caregiver immediately upon a client's request. The fit between a caregiver and senior is a very tricky dynamic.
- Long lead time from traditional channels. Most companies recruit caregivers through Indeed, Craigslist and newspaper ads. It takes on average 7 days and $400 to hire a new caregiver.
Coming Up with an Hypothesis
Traditional recruiting channels like Indeed and Craigslist do generate qualified applicants. But the manual work involved in filtering candidates for the job can take up to 30% of the operation department's time. Those manual work are mundane tasks such as
- making sure the candidate has proper training certificate
- has a valid police background check
- is available to work within a certain time frame
- isn't allergic to cats (if applicable)
- and much more
So we began to wonder - wouldn't it be cool if there's a way for home care companies to receive qualified candidates fast? So that they can serve clients wherever they want and whenever they want.
Although we thought we had come up with a clear value proposition, we weren't sure if it is something that the home care companies wanted. Also we were wondering if it's something they were willing to pay. Second, how are we going to supply home care companies with qualified candidates fast? What can we do differently to get fast turnaround than traditional recruiting channels? If we can use very little effort to find answers to these two questions, we'd be able to confidently say when we should start writing code.
1. Validating Hypothesis
To see if this is something that the home care companies will find useful, we want to be able to test it with a solution that is cheap, fast and low-fidelity (so we don't become attached to it). We thought about using email as a way to present our candidates because it is what we're all familiar with and requires us to not build anything. So far, we haven't define what "fast" means since we have no idea how much faster our service is than traditional hiring model, so we decided to really work on quantifying turnaround time in the next step when we work on differentiating features.
We decided to use the following steps to validate our hypothesis:
- get five home care companies onboard to pilot test our service
- set up an online form to capture hiring inquiry data
- send email with qualified candidates based on company's inquiry form data (fake candidates)
For us to move on to the next stage, the pilot companies will:
- tell us that they find this useful and would like to use it to fill real shift
- give us their credit card (although we'd offer than first 2 hires on us)
After one week of testing with five home care companies, all of them have said that they found receiving the candidates' email very helpful, in a way that they feel like all they have to do next is to call the candidates and start scheduling phone interviews. All but one company didn't give us their credit card information, so we decided this was strong enough of an indicator that this is something companies would pay money for.
Now that we've validated that home care companies definitely want to receive qualified candidates fast, the next questions is how are we going to pull it off? What can we do that is different than what traditional channels work? Before coming up with any ideas, we've established the fact that the idea has to be fast and generates qualified candidates.
One idea went off like a lightbulb in our head. When we were providing home care services ourselves, we'd give out our personal cellphone numbers and the caregivers would text us when they need directions to a client's home, notify us of a schedule change or provide general feedback. The caregivers as a subset workforce is very familiar with using text messages as a means to manage their work. What if we send out available shift information through text messages? We then started working on breaking down this idea into actionable steps:
- recruit caregivers. If we're focusing on best matches, we'd need a pool of caregivers available to maximize the chances of always finding a fit. Our goal was to recruit 100 caregivers in a month.
- send outbound text messages. Can we do this without writing too many lines of code? After some research, we realized that Google spreadsheet has the capability to send text messages via twilio api just with a few lines of code.
- respond to text messages. To do this, we used Front to receive text messages and reply to responses for each caregiver.
After one month, we were able to recruit 200 caregivers. By definition, a successful recruit means that the caregiver has filled out an online form where we collect their personal information and has uploaded a copy of their resume, valid police check, caregiver certificate and other relevant documents.
After testing this solution with 6 real shifts from our pilot companies, we were able to get around 50% of the responses within 3 hours from the message is sent. In other words, if 100 text messages are sent out at 3pm, we'd have 50 responses by 6pm and around 20% of that would be "YES" which we will then forward those candidates to the home care companies.
We've accomplished two things so far: 1) validated that this is a concrete need for home care companies, 2) figured out an execution strategy that can generate qualified candidates fast. This solution has been working well for the first 15 hiring requests but we've been learning there're the following drawbacks:
- companies want the ability to manage candidates based on each shift as emails can easily go out of control
- aside from taking up too much time, manually filtering for candidates and sending out text messages made us very prone to errors