top of page
Artboard 1_3x.png

Crafting a Seamless Search Experience

The redesign of the Demyst Labs platform in 2018 had a tremendous impact on the Demyst user experience. However, the the new platform opened the gates to new types of users with unique challenges that we needed to solve for. 

demyst_preview.png

The Opportunity

In April 2019, DemystData had just launched a suite of data discovery, analytics, and governance tools united under the Demyst Marketplace. The heart of the marketplace was a searchable product catalog, with a real-time interface to sample files, dictionaries and fill statistics. At the same time, Demyst was rapidly scaling the acquisition of new data products and in just a few months featured over 500 products with 100,000 attributes (individual columns of data).

Despite the success of the positive reception of the catalog, touchpoints between our users and customer success team remained high, and users struggled to find the products they needed.

To learn more about the the Demyst Marketplace, read the case study DemystData | Designing for Data Analytics Teams.

Start

Insights and Core Challenges

We conducted a series of interviews with a diverse group of enterprise and marketplace users. Our goal was to understand what information users were searching for on our platform, and what we could do better to bring that information front and center.

 

In addition to interviewing users face to face, we dug into our event tracking data and google analytics to really understand what key terms were really searching for. The following were some key design questions that emerged:

  • How might we display information in a way that aligns with what users are searching for? 

    • Most users search for terms related to specific attributes, not specific products. Users don’t know what products they are looking for. They know their business problem and what fields of data they might need to solve it.

  • How might we make the search experience more intuitive for users based on common behavior?

    • We found that attribute performance statistics and price were the deciding factors when choosing between products.

  • How might we leverage our marketing content to educate and drive user activity? 

    • User searches were highly influenced by our marketing and blog content. For example, in the weeks following an announcement about a new Property related source, we saw a spike in terms like “landgrid” and “square footage”.

Competitor Research

As inspiration for my design, I decided to research how travel websites structured their search experience. I felt that the user experience of buying a flight would be very similar to that of buying a data product. 


Most people looking for a flight don’t know if they want to fly Delta or JetBlue. They do know that they want to fly from New York to Los Angeles on Monday, May 2nd and their budget is about $600. We found the same to be true of data buyers. 

Kayak Notes.png

Concept Ideation

Concepts were developed through a series of rapid prototypes, with feedback sessions between each iteration. I was responsible for the end to end design, content strategy, and search experience.

Key Ideas

Capturing User Intent.

By putting advanced information and CTAs in the expanded accordion view, we were able to track which attributes users were most interested in.

Pagination.

Displaying our catalog at the attribute level instead of the product level meant displaying almost 200X more cards. To do so without slowing down website performance, we decided to add pagination logic to our page. This also had the effect of conveying the breadth of our catalog.

Purposeful Filters.

We made filters like Match Rate and Price more prominent because those were the two most important factors to users when choosing a product. We also decided to group our products into price ranges to give users control over their budget without limiting the breadth of their search.

Relevant CTAs.

After a few iterations, feedback showed that we were skipping too many steps between exploration and data access. In most scenarios, before purchasing a data source, users would want to explore more information about the product and get a few samples before making that commitment. In the end, we decided on a ‘Learn More’ CTA that led to our existing toolbox where users could access sample data, dictionaries, and learn about the attribute source.

Lead Generation.

To keep the momentum going from ramping up our marketing content, we added a form to capture user emails. This allowed us to capture users that didn’t convert right away and target them with relevant content.

Final Concepts

Kayak Final.png

More Case Studies

bottom of page