MOZ CONTENT SEARCH
The content search feature was our second main feature for Moz Content. While the content audit helped users understand how their own content was performing, the content search allowed them to discover the top content around the web by topic and get key insights into how it was performing. This in turn helped inform decisions they were making on the forefront of their content strategy. Super valuable.
DELIVERABLES I WORKED ON
SKETCHES | WIREFRAMES | USER FLOWS | DESIGN COMPS | RESEARCH
The Filter Problem
We came across some problems with the early version of our content search, specifically with our search filtering interaction. Our first version had all the possible filters shown before a query was made. As we learned through some research, this encouraged the user to add filters before seeing the base of their query. This severely limited the results of their search or returned none at all, giving the perception the feature was not very useful.
Our first step to fix this was to initially hide the filter in an "advanced filter" drawer. Once the user had returned results from their search, the drawer would expand to reveal the filters. This encouraged them to further refine their search if the initial results weren't specific enough.
While this greatly reduced the amount of help tickets we had centered around the results being weak, it still felt too complex for a filtering interaction and I knew we could simplify it to reduce cognitive overload. After testing several designs, we found the filtering could be accomplished effectively through two simple drop downs pre-populated with the broadest option.
We replaced the advanced search drawer with a guide for power users wanting to use search operators to refine their searches further.
My favorite interaction
Each piece of content within the results of a search had really great data about its performance. A valuable piece of this data was found in the shares for each social network. This gave the user insight into what social networks the content was getting the most shares on.
As we continued to refine our search feature, our system got smart enough to be able to define how a piece of content was performing on a specific social network in relation to other content with its same attributes. We created a scale showing if the content was performing in the top 1%, top 10%, top 25% or last 75%. We were stoked to provide this added value and needed an intuitive way to communicate this to our users. I wanted to capture the emotion involved when a users content is performing well. It's a moment to celebrate. I worked with our creative team to create emoji's of Moz's mascot Roger, which would represent each of the categories within the scale.
There was great response to this interaction and it was fun to bring a delightful moment into the product, which added personality and emotion to the experience. One of the last details we were debating among our product team was surfacing the emoji's in place of the color indicators. Not only would this be a more immediate indicator of how the content did on that network without having prior knowledge of the scale, but it would also be better from an accessibility standpoint allowing those who are color blind to understand the sentiment through visuals.
The latest version of the content search feature had added a lot of value from where we began. It surfaced related topics to help you refine your search as well as a summary of the topics performance across social networks and a timeline for how the topic was trending across the web.
Our content search feature became a top tool for content discovery and was written about by several industry blogs regarding it when paired with our content audit as one of the best tools for content marketing. Here are some things they had to say:
What I Learned
I learned a lot about distilling complex data to understand users needs and provide it in a way they get value from. Dealing with individual search results and how they are best consumed was a challenging problem to solve and I am really happy with the way it took shape. I also learned a lot about the tendencies of users to over filter a search if you let them. The work on refining this flow was challenging and it was rewarding seeing the affect of the refinements directly through a reduction of help tickets.