When working with modern technologies, the use of legacy design principles will only yield short-term benefits. The end result may satisfy purely functional needs, but it often falls short of truly addressing human behaviors and desires. For this reason, many organizations are turning to data-driven design. By applying analytics to how users navigate systems and consume data, businesses can derive valuable design-oriented insights.

Further, by applying artificial intelligence and advanced forms of machine learning to data-driven design, businesses can quickly identify the intersection of feasible, desirable and viable innovation options. Considering the market advantages made possible with a balanced innovation approach, it’s critical for companies to embrace minimum viable products (MVPs) to maximize the return on innovation-focused investments. For that, a sophisticated technology design toolkit is a must.

Businesses that adopt data-driven tools can also compress innovation timeframes. Although design thinkers are skilled at using empathy to gain insight into users and their needs, their assumptions often increase the elapsed time from ideation to solution optimization because of the multiple iterations that effective innovation requires.  Data-driven tools can help validate assumptions more quickly.

Getting an Automated Read on Human Needs

It is not enough to merely design the MVP effectively. Equally important is continuously keeping customers at the center of product design and innovation so that real-time user feedback truly drives continuous updates to product, platform or solution design.  Natively-digital organizations like Salesforce, Alphabet and Amazon do just that. However, when it comes to digital immigrants, the progress is always incremental.

What follows are three examples of how data-driven approaches can facilitate human-centric design.

  • Case #1: Strengthening the enterprise architecture by automating the identification of design patterns

    We worked with a UK-based public transportation company that was grappling with its aging Oracle integration platform. At capacity and facing supportability issues, the system itself wasn’t scalable, and the invoice processing system could address only one invoice at a time. The result was a significant backlog.

    We divided the project into five work streams and conducted design workshops for each. The goal was to develop a clear roadmap for merging the five streams into two during the execution phase. We applied analytics to the company’s vast amount of user consumption data to derive many insights geared toward improving the enterprise architecture design.   

    Applying design-thinking principles and data-driven design, we evolved MVPs for each of the work streams, which led to a standardized service-oriented architecture and business process management integration platform. Additionally, by feeding the design constraints for user needs such as time and data quality to machine learning software, we generated thousands of potential designs for components that a designer can choose from and then further develop for future releases/extensions.

    By not just designing for the problem at hand but also continuously keeping the design aligned with customers’ consumption patterns, we guaranteed a more meaningful return on investment for the customer.

  • Case #2: The limits of empathy and the shortcomings of traditional human-centered thinking

    Does asking what one human being is thinking or feeling as part of a design-thinking exercise actually boost insights into what others are thinking and feeling? Research indicates the answer is no. To create a meaningful design, it’s absolutely critical to get the human perspective. But it isn’t foolproof.

    An example is a cloud computing company whose service portal was hampered by an inconsistent user interface (UI), a fractured information architecture and poor navigation features. Although the portal was known for its robust, configurable features and domain workflow, the company’s customers wanted a more consistent, innovative and intuitive user experience across every scheduled six-month product release.

    We kickstarted this engagement by auditing the company’s applications and hosting design-thinking workshops. We proposed an employee-centric approach to designing enterprise service management applications by developing guiding principles and crafting a design theme suitable for all IT applications. We centralized the information, making it contextual and easy to find, which facilitated early adoption of the platform. We also introduced global smart search, personalized dashboards and chatbots to the platform, as well as a knowledge base that provides additional insights such as “top-rated” and “most viewed” content.

    In addition to developing scalable and responsive designs for mobile and desktop options, we defined a living UI toolkit so designers and engineers can reuse the design style of components, ensuring a consistent experience across applications. This enabled the company to transcend users’ current perspectives, and create a fit-for-purpose UI. The toolkit made it possible for designers to consistently maintain a satisfying experience throughout the life of the applications.

  • Case #3: Designing with analytics in mind

    Applying analytics to design is the only way to derive useful and scalable insights. To do this, designers need to pay attention to data management.

    For instance, one of our clients – a U.S.-based utility that serves approximately 1.1 million electric and 790,000 natural gas users – needed to upgrade its legacy systems. The company wanted to provide its customers with better energy options and engage them more effectively by creating hyper-personalized experiences that addressed their unique energy consumption needs.

    The utility also wanted to develop a system in which data drives optimal business decisions. Along with the use of smart devices, the company realized that analyzing customer data could reduce customer defections to emerging competitors.

    We applied analytics to predict and handle the utility’s increasing call volume – estimated at four million calls annually – as well as inform customers of impending outages and their resolution by text or e-mail. We also proposed a two-track approach, first creating an intelligence platform for the existing IT landscape and then using a business track to create a use case inventory and prioritization framework.

    Lastly, we implemented the SAP Master Data Governance module. Solutions were designed to provide customers with self-service capabilities, such as bill explanations and payments, financial assistance requests, service interruption updates and service event planning, across multiple channels, including online, mobile and customer care centers. We also designed solutions to help customer service agents make better real-time decisions.

Up Next: The Emotional Interface

Measuring the efficiency of effective designs remains a work in progress. In the above examples, continuous user feedback during the nonlinear cycle of the design-develop-test process helped reinforce both the design patterns gleaned from system usage and design consistency of all components/applications. Doing so ensured maximum business value.

But design effectiveness is not measured by merely assessing whether the resulting product or service has solved the intended problem. Design can only be deemed effective if it generates a positive emotional reaction from users every time they use the product or the solution.  Understanding this through focus group discussions and online surveys is only one important first step.

Perhaps the only way to truly know how humans feel when they interact with systems is to embed electrodes into their hypothalamus and measure their emotional response – that’s a topic, though, for another post.

Anbu Muppidathi

Anbu Muppidathi

Anbu Muppidathi is a member of Cognizant’s executive leadership team and is a senior leader in the company’s Digital Systems and Technology... Read more