On her bus ride into the office, Anne is casually browsing Instagram on her phone and sees a dress she loves from Free People in her feed. Later on in the day, Anne’s boss is in a meeting and she’s bored at work so she decides to visit freepeople.com to search for that dress for an upcoming wedding. On her way home, she decides to stop by the Free People store where she tries on the dress, decides it’s a winner and makes the purchase.
The goal of the Revmetrix platform is to be able to show the near real-time journey of a customer through the entire shopping experience so retail analysts, buyers, and marketers can better understand how segments of people make purchases. For those professions, the current process to get that information is slow, sometimes taking weeks and segmenting through various buying channels to a unique shopper is impossible.
"If your product was a person, who would it be?" The general consensus with the Revmetrix team after some brainstorming was IBM Watson’s meets Nate Silver meets Squarespace. Basically, super smart but pretty. We used that finding through the rest of the design phase.
The Revmetrix team was hard at work for 9 months working to get the backend of their MVP set up before they were ready to add in an interface. The product manager created product wireframes he created based on the requests of retail buyers and analysts. From there, we worked together to refine what information was to be shown and how it would be displayed.
Since the goal was to get an MVP out as soon as possible, the priority was quickly designing an easily configurable interface that would be flexible enough to allow the addition (and subtraction) of features and information in the near future with less attention placed on the visual design.
The functionality of the MVP was thoughtfully simple and included only precisely what we predicted the users would need to be successful. Focus was also placed on how to take complex queries and make them super easy to comprehend. To avoid the problem of information overload, we displayed only the bare minimum of information needed for the buyer or analyst to make informed decisions.
We chose a side navigation bar for quick context switching between data sets and added in bookmarking for users to find commonly called queries. We determined the most common action the user would take is “creating an insight” or the creation of a custom query selecting specific segments of the users base according to how they were shopping (online, offline or a combination of both.) The “create a query” button was prominently placed at the top right hand side for easy access.
The marketing site was designed to appeal to the retail buyers and analysts and to humanize the consumer-side shopping journey. So, on the homepage, we decided to illustrate the process of an individual user throughout the “omni-channel” buying process.
The photoshoot took place on muggy on summer day in DC, where I followed our model around shopping in store and occasionally then checking to see if she could find a better price online. (Who hasn’t done that before?). Finally, finishing up transactions both at home and at her office.
Through the remainder of the site, we wanted to clearly articulate the value proposition of the product to potential users in language they would understand.
The development of the site was not completed internally and changes have been made since the design phase.