Care.com is a marketplace for families and caregivers to match based on things such as pay, schedule, and quality of care. I worked on a team focused on the caregiver experience, focused on helping them improve their chances of finding a job on our platform. My team was solely responsible for the entire caregiver experience (exactly half of the product), one which has largely been neglected for many years.
Using the resources from our Consumer insights team in addition to talking to some users in person, we were able to identify some pretty glaring pain points for both seekers (families) and providers (caregivers). Some major complaints were:
After having established a regular cadence of interviewing users, reading product reviews, and listening to feedback from our Consumer Insights team, we saw a major area of opportunity for improving the user experience of our product. We determined that if we:
... then we would significantly improve the ease at which people find suitable matches on the site, and thereby increase our match rate, NPS scores, and overall revenue. Here are our hypotheses, based on real user problems:
When looking for inspiration, it is useful to try to find competitors that are solving a very similar problem to yours.
The idea is to take what they've done, and do it better.
We looked at a number of competitors, including Thumbtack, Urban Sitter, Sitter City, and Indeed. All of these platforms have some sort of calendar tool, using a concept of "general availability". Each of these platforms give very little capability for their users to be specific about when they are available. For example, Thumbtack only allows their users to select 1 time slot attributed to all of the selected days of the week. Urbansitter, on the other hand, allows you to select different time slots on different days, but the time slots are only "AM, PM, and Evening".
When it came to the point of user testing, we focused on finding answers to specific hypotheses. For example, we hypothesized that users preferred to input weekly availability using broad time-slots because of the perceived ease, speed, and simplicity of the form. I made a prototype showing 4 different ways that caregivers could input their availability, and discovered, to my surprise, that providers preferred to input very specific, complex availability times.
We did a similar test for seekers. This time, instead of inputing their own availability, they were shown what the search filters might look like, combined with how the caregiver profiles might show available times. Again, to our surprise, seekers preferred the more specific, complex availability times.
After testing many of our hypotheses on usertesting.com, we came up with a design that we feel will satisfy the needs of all of our caregivers, while also helping to improve many of the problems that seekers have regarding finding available caregivers. This design features: