Rating a Professor
Lead Product Designer • March - May 2020
RateMyProfessors.com is the largest online destination for professor ratings. It allows college and university students to anonymously rate their professors and campuses of American, Canadian and United Kingdom Institutions. RateMyProfessors.com was founded in 1999 and the site includes 8000+ schools, 1.7 million professors and over 19 million ratings.
My RoleI co-owned product strategy, research, wireframing, visual design and prototyping. I worked closely with a product manager, the engineering team, and the revenue teams to understand the multiple facets of our current experience. In addition, we procured the expertise of a community manager to voice nuanced observations directly from our active user base.
DEFINING THE PROBLEMA steady decrease in rating completion
Rating a professor is an essential user action that helps maintain traffic on RateMyProfessors.com and is one of the most commonly used actions following searching a professor and viewing a professor’s ratings. The Rate a Professor page gets on average 330k unique visitors a month.
Early in 2020, the drop off rate was ~65%. This led us to believe that a lot of users were arriving at the Rate a Professor page, but not completing a rating. Since the credibility of RateMyProfessors.com relies heavily on having recent and quality professor ratings, a high drop off rate on the Rate a Professor page could begin to negatively impact the end user’s experience and dilute the credibility that our site holds.
I suspected that the existing design of the Rate a Professor page was causing a lot of friction for users and greatly attributed to the high number of drop offs.
MEASURING SUCCESSI hypothesized that by addressing major pain points and creating a more intuitive, dare I say fun experience, more students would complete the process of rating a professor and thereby improving the overall health of RateMyProfessors.com.
Success would look like:
Reduction in reported issues
YoY increase in ratings being left
DISCOVERY
Obstacles in the path
To get a deeper understanding of the decline in ratings, my team conducted an unmoderated usability study observing over 100 recordings of users landing on the Rate a Professor page. We chose this method because we wanted to leverage the high traffic RateMyProfessors.com receives on a recurring basis. We didn’t feel the need for nuanced 1-1 learnings of an on-site qualitative interview. Using a screen recording tool, we could get quickly get a representative sample that would inform general behaviours and trends, which we organized by affinity mapping.
Scrolling:
We noticed a lot of users scrolling the entirety of the page when they landed. We hypothesized that students wanted to scan all of the questions before committing to beginning a rating. At the written part of the rating, we noticed students again scrolling up to refer to their previous answers and sometimes making adjustments.
Bouncing:
Nearly half of the users who landed on this page bounced immediately.
Upon further investigation, we found out that when a user landed on a professor’s page that didn’t have a rating yet, instead of an empty state, users would be directed to the Rate a Professor page - a flow that we found extremely disorienting.
Errors:
Of the users who went through the rating process, many received an error when trying to submit the form.
We hypothesized many pathways to confusion & unnecessary friction including captcha validation, agreement to terms, and the sign-up experience that was built in to the Rate a Professor page.
Confirming the Professor:
We observed many users navigating away from the Rate a Professor page to look up a professor’s profile, then navigate back. We hypothesized that students were trying to confirm that they were rating the correct professor as some professor names are common or have multiple profiles if they’ve taught at more than one school.
Using a customer journey map, I brainstormed opportunities to our problems. The discovery process led me to believe that the existing Rate a Professor experience could be improved. I set on the spiritual journey of Marie Kondo-ing the Rate a Professor experience: clearing the path, reducing obstacles and hopefully, sparking a little joy.
EXPLORING SOLUTIONS
Less is more
To minimize scrolling, I moved the site guidelines from the top of the page to above the written part of the rating, where it was most relevant. I condensed it all in to a tidy accordion that would automatically be collapsed for return-raters.
For professors that didn’t have ratings yet, I created a simple empty state to encourage intentionality of starting a rating. Users would now have to click a CTA to rate a professor rather than being dropped in the rating experience without context. A week after making this update, we saw a 36% increase in completed ratings.
To minimize the number of errors at submitting a rating, I moved sign up to after rating submission was confirmed and added value propositions to give users a little push. It was important to reinforce reward for input and remind users of the benefits of having an account with RateMyProfessors.com the next time they thought about looking up a professor.
By pinning the professor’s essential information to the top of the rating page, I hoped to save users from having to navigate away to confirm they had the right professor.
Finally, to spark a little *~joy~*, I created a set of colourful selectors that clearly indicated the direction of the scales as some went from positive to negative while others went from negative to positive.
NEXT STEPSPruning variants
Historically, the freeform experience of adding a course code has resulted in high variance, making the experience of filtering ratings by course code less than efficient. For example, a course named PSYCHOLOGY1000 may be entered in many different ways; PSYCH1000, PSY1000, etc. One way I have been thinking about reducing variation is creating a path that encourages users to search and select from a list of existing course codes and only adding a new course code if it didn’t exist. This experience could be implemented with minimal effort but may have a significant impact on the filtered search experience moving forward.
Like this but a little less dramatic
Looking aheadSpreading joy
Lead Product Designer • January - April 2021
As we continue to roll out improvements to RateMyProfessors.com, we have recently turned our attention to the mobile app experience, where we are scaling our learnings across platforms. Throughout my process, I have ensured that my explorations were consistent with RateMyProfessor.com’s global design language, designing components that are responsive and adaptive to Android or iOS.
