Client File: IBM

In 2015 IBM was endeavoring to establish a design language and guidelines for data visualization. General Manager of Design Phil Gilbert contacted Accurat to design, develop, formalize, and deploy a unifying system that his team of 1500 designers would adopt.

We spent a full year conducting field research at IBM’s Austin headquarters. During this time, we collaborated with different teams to prototype solutions for real products in different lines of business: AI analytics, IT security, and drug discovery, to name a few.

Meanwhile, we dove into IBM's archives, familiarizing ourselves with the company's vast visual heritage. The work of legendary staffers served as both inspiration and a starting point. Homing in on recurring visual motifs, we designed on–brand data visualization templates for every use case.

While conducting research for Design Language, Accurat interviewed designers working on Watson Analytics: a data analysis and visualization application powered by IBM’s famous AI technology.

Soon after, the Watson Analytics team reconnected with Accurat to weigh in on a case that they hadn’t yet been able to crack. Even Watson can’t answer some questions with certainty. For these queries, how could designers visualize predictions with precision? Naturally, our response would need to align with the guidelines we’d developed ourselves.

Accurat zeroed in on the grey areas of forecasting -- the margins of error -- and determined the best ways to visualize uncertainty.  We put together a framework to pass on to IBM: a handbook that included research-backed reporting, design explorations, possible solutions for IBM, and references.

IBM Security develops intelligent solutions and services for businesses to combat cybercrime. QRadar is their top-selling, award-winning SIEM (security intelligence enterprise management) product.

While working on Design Language, Accurat also met with members of the QRadar team. Later, they contacted us to design an advanced visualization environment: a feature QRadar had previously lacked. Whereas we had previously acted as advisors, with IBM Security we took on new responsibilities as developers.  

We designed and wrote production-ready code to build an advanced visualization interface. The new version displays global dangers in real-time, along with drill down features that expedite investigation, facilitate robust analysis, and equip employees with intel to take more proactive action.

Accurat continued to meet and work with other departments at IBM. The next project represented a shift in scope: an experiment that evinced IBM’s growing trust in Accurat’s capabilities.

Think Summits are local meetups that give technical education and business guidance to ticketed guests. In 2019, the newly opened IBM Studios Milano was set to host the conference.

Keeping with the year's theme of “Republic of Ideas,” IBM Italy CMO Luca Altieri wanted to promote the space as a new, permanent center for learning in the cosmopolitan city. He contacted Accurat to come up with a creative, interactive solution.

Accurat designed an experience that paired a public display in Milan’s tech center, Piazza Gae Aulenti, with a personalized keepsake to remember the event. Before the conference, guests were invited to create a digital “ID card” in a custom app that generated a design based on their responses. At the event, anonymized ID cards were pieced together on a massive outdoor LED screen to resemble a flag for the “Republic of Ideas.” Later, a follow-up questionnaire updated ID cards to remind guests of the recent experience and gauge how their opinions had transformed because of it.

Watson Studio is a cloud-based software that helps data scientists and analysts build models to forecast and optimize business outcomes. In 2019, they felt as though their product was falling short of IBM’s high standards for innovation. By this time, Accurat’s name had spread through IBM’s grapevine. They contacted us, asking for input on two questions:

1. How can we extract information from unstructured datasets without manually combing through them?

2. How do we monitor the training of a machine learning model?

Similar to the work prepared for Watson Analytics, we compiled design ideas and prototypes for an online repository that presented research and ideas for solutions in documents, demos, and prototypes. 

IBM Services is the company’s $17B consulting arm. They are an elite team of strategists who work at the intersection of business and technology, solving for problems like  business continuity, remote workforces, and virtual client engagement for the world’s largest companies.

Consultants are also responsible for introducing software like Watson Analytics and Studio to client workplaces.

Because these products rely on data visualization, it’s crucial that representatives understand how it works. Managing Partner Jesus Mantas felt as though his team could use a refresher in 2019, and knowing that Accurat was responsible for developing Design Language, he asked us to work with his team next.

We built a curriculum comprised of three modules to give IBM Services employees a crash course on data visualization and its relevance to AI. Why? As stated in section 1.5, “Data visualization can provide the user with a way to access the black box of AI processes.”

When the Brand Experience and Design group took ownership of Design Language and relaunched it in 2018, Executive Creative Director Mike Abbink asked us to update guidelines for all IBM experiences—including those beyond digital. 

We worked with Abbink’s Brand Experience team to revamp material for Carbon: an open source website that consolidates resources for typography, photography, illustration, and all other facets of IBM’s visual identity.  In addition to defining five criteria for effective dataviz (understandable, essential, impactful, consistent, and contextual), we also developed the web components for all chart templates, with integrated how–tos to facilitate easy adoption.