Rakuten Design Challenge

The Challenge

Design a website market research and analytics dashboard for a new product at Rakuten, Slice Intelligence. Its helps marketing or data analysts better understand how users are engaging with e-commerce by managing receipts in your email inbox.

I was given 48 hours and the creative freedom to deliver the product as I see fit. Just like every other design, I move through three main phases: Inspiration, Ideation, and Implementation.


The Process

@JoseCaballer @theskoolrocks has a great post about anchors in UX design and how they manage the flow of information that guides collaborative decision making in a harmonious way. These anchors are important to establish early in the process and that starts by learning about the users and their lives. Doing this saves the team time, and sanity, leaving more for making delightful products that people actually want to use. 

Learning directly from the people we’re designing for is core to human- centered design. Through interviews and observations, we seek to better understand the whole person, as well as the contexts in which they live. We often start seeing patterns after about eight interviews. 


One particularly interesting conversation and observation is often more valuable than 100 shallow ones.

Research and Discovery

The current Slice Intelligence dashboard retrieves the purchase data from receipts in users inbox who have linked their emails to the Slice App and unroll.me (which is their email subscription management tool). Some of the products/reports currently available focus on providing volumetric insights ($, units sold, buyers) while others show email volume trends (shipping, order, confirmation, payment, etc). 



Competitive Audit

A competitive audit was conducted to assess product features, prioritize user pain points and identify new opportunities to explore the data collected by the Slice . Because I already know what kind of tool I am designing towards, it was helpful to conduct this first to help narrow my target users. It will also provoke questions that will be helpful to ask during user interviews.


Adobe Test and Target is an A/B, multi-variate testing platform which Adobe acquired as part of the Omniture platform in 2009. It is now part of the Adobe Marketing Cloud. It offers tight integration with Adobe analytics and content management products.


The Pros

  • It is simple to use and gets the job done fast
  • A/B & Multivariate testing capabilities and features
  • The level of product support
  • Rules-based targeting
  • Recommendations and cross-selling
  • Automated behavioral targeting and Multivariate testing
  • Drag and drop interface for setting up tests is great for less experienced members of the team.

The Cons

  • Adobe Test & Target comes at a premium price
  • Usability for non technical users
  • To build out specific segments required direct knowledge of how to write javascript
  • Does not even integrate well with the suite even though it is from the same product line


The Pros

  • Building customized reports is very easy
  • Allows marketers and analysts to construct very specific user segments and see 100% statistically valid trends and datasets.
  • Can show real time traffic to the site, useful to see if commercials, ads or eblasts are directly generating interest.

The Cons

  • missing some key metrics and reports for retail/ecommerce brand (product views, add to cart, abandoned products, cross sold product basket analysis, etc)
  • You can not share segments across the organization or groups
  • Creating goals and conversion funnels can be difficult for novice users

Google Analytics Premium is an enterprise-level analytics solution that includes a full service suite of features offering a singular view of the consumer across platforms and devices. It also provides full integration with the Google stack, data-driven attribution, and the ability to optimize site content through content experiments.



I have a few friends that work in online marketing, so I reached out to conduct some user research. While I wasn't able to talk to eight users, I did get the chance to interview five marketing and product analysts. Some of the user profiles overlapped so I merged them into 3 personas and chose a primary user persona to focus on the features that would benefit them the most with Slice's data. 


Everyone seemed to want the same things - to be automated, easy to find new leads (stories), and the ability to create segments. The range of industry professionals I interviewed varied in skill level, but most agreed that the UI in current industry platforms was difficult to navigate for novice users so I wanted to focus on the interaction and information architecture for this design.


Information Architecture

During one of the interviews I was told that a lot of her inspiration came from Google Trends because "there is a diverse range of content that always provides something relevant," so I wanted to explore what the source of Google Trends Information Architecture was presenting in its design.

there is a diverse range of content that always provided something relevant

Whether you are looking to find new areas of growth in your product portfolio or new ideas for a marketing campaign, this tool provides insight into various topics across the web utilizing sources like news, social media and traffic through a simple range of categories. It also allows you to add your own filters to narrow your search. 

What I found fascinating was the fullness in the experience of content when using this tool. No matter what I searched for, google took the vast amounts of data online and transformed it into valuable insights from the categories provided. It looks something like this:

  • Most relevant
  • Interest over time
  • Interest by region
  • Trend search 
  • Related topics 

I wanted to validate whether this architecture would also apply to the data from slice and provide the same value. After some analysis, this became the underlying information architecture for my design.


Execution (Wireframing and Prototyping)

I use responsive design patterns to build out hi-fi wireframes in Sketch that enable you to test your projects with real users on real devices. 


Search Feature Design

Segment Feature Design

Search was the other focus, finding deeper relationships through data trends.

This was the core feature identified through competitive analysis and user interviews.