IBM Design, 2015

Collaborator: Agile team of developers, designers and product managers

Watson Analytics is a new mobile experience for people to make business decisions with data and tell compelling stories about those decisions. The application puts data in the hands of decision-makers rather than data scientists, making it easier for people to share their data-driven insights with their collaborators.

Erin's contributions to the product include: conducting generative user research and conceiving aspects of the user experience for assembling a view of data, pinning data visualizations, and storytelling.

Welcome Screen

One of our primary research insights is that people rarely start their analysis from a blank page. We designed the experience so that people can reuse pieces or entire documents of their past work, start by writing their business intention, or begin with their raw data.

Create a view of data

Select a page format

Add a pinned data visualization

The user drags and drops a pinned data visualization saved in his pinbar to a new view.

Smart guides

Smart guides help a user who knows his data and knows his business but is not an expert in visual design align and size elements.

Add a new page

Create a new data visualization

A user can create a new data visualization by dragging a data column onto their canvas. Watson Analytics automatically generates a recommended visualization that matches the user's chosen styles.

Select additional column of data

Add column of data to existing visualization

When a user adds another column of data to an existing visualization, Watson Analytics automatically updates it to reflect the new analysis. For instance when a user adds location-based data as shown here, Watson Analytics transforms the sales total into a visualization of sales totals displayed by state. The user can further refine the data visualization, giving it a different form if he does not want a map.

Experience Overview

This experience overview was developed early in the development of Watson Analytics to identify and define the aspects of the user's experience our team would work on. We subdivided the work further into three parts indicated here: modeling data, building a data application, and creating and using a pinned data visualization.

Experience Overview (draft)

Here is an earlier version of the experience overview.