Harnessing Data for Human-Centered Design

The need for Humans to be connected to other Humans is as basic and essential as the need for Food, Water, and Shelter. Many neurologists and human-computer Interaction experts have studied the evolutionary aspects of the relationship between social connections and the overall well-being of Humans in various historical forms of prehistoric and civil society structures. In the Modern World, this aspect has translated into most of us being connected to the Internet as an essential part of our day-to-day lives. According to Statista’s survey data reports as of January 2024, there were 5.35 billion internet users worldwide, which amounted to 66.2 percent of the global population. With such an enormous digital presence, users of the internet have a major influence on Social Behavior and Economies around the world. With the consistent growth of Internet usage each year, Data and Design practitioners have an increased focus on designing digital experiences that are deeply user-centric. This has resulted in Websites, Applications, and other Digital Experiences that deeply cater to the needs of the user based on who they are.

Data Informed Human Centered Design

With almost two-thirds of the world population having a Digital presence, the Internet experiences they use must be designed with the “User” as the central focal point. The Design community has embraced various forms of Data-driven design practices over the last few decades which has resulted in UX ( User experiences ) that are accessible, equitable, and empathy-driven. These Data-driven Design practices can be broadly grouped into Attitudinal-based and Behavioral-based. Attitudinal User data research involves gathering user needs and preferences based on self-reported methods such as surveys. Behavioral User Data research involves studying digital user behavior to identify navigation patterns across digital experiences and infer insights into their digital consumption patterns. With learnings combined from these two methods, artifacts such as Empathy Maps and Journey Maps can be combined to effectively understand and summarize complex user needs and behavior. In this article, I will be explaining the concepts of Journey Analytics, Journey Mapping, and Journey Orchestration as it relates to Data-driven Design. Though the concept of Journey mapping has been widely popularized by marketing teams in the past for their applications on Conversion optimization, the original concept still remains a critical aspect for Experience design and Human-computer interaction disciplines. Throughout this article, we will be focusing on the Computing and Data mining aspects of Journey analytics that enable delivering hyper-personalized user experiences in a large scale.

Journey Mapping

Journey mapping is the process of identifying who your users are and mapping the various touch points through which the users interact with your product/ user experience. Journey Mapping involves identifying the various journeys the user can take to reach their goal and mapping out attitudinal and behavioral data points in each stage of their journey to produce a visual map that lays out everything there is to know about the user’s emotions and actions. In its simplest form, Journey mapping exercises are typically conducted to ideate the journeys a user can theoretically take while navigating user experiences.

While this approach may work for nudging the users towards a singular goal, modern large-scale UX optimization efforts aim to understand the needs and preferences of the users and help them accomplish their tasks in shorter and more effective journeys. In reality, user journeys can be complex and unique to every individual and highly deviate from the expected user path. It is also important to understand that journeys happen on different devices and can span several days. Exploratory journey analyses are conducted regularly to understand the various user journeys to identify user habits to fine-tune and revise the Journey maps.

For example, a user using YouTube can begin their content viewing journey in their mobile device and eventually switch to their TV. They may also pause and continue it several days later. Aggregating such unique user paths for millions of real-time users requires large-scale data exploratory analyses and calls for dynamic journey maps aided by insights derived from Journey analytics.

Journey Analytics

Each phase of Journey mapping needs to be fortified with metrics that effectively measure the Attitudinal and Behavioral data points that capture the user’s intent, emotions, and actions. This phase is commonly referred to as Journey Analytics. Journey Analytics setup typically begins with identifying goals, pain points, and benchmarks for each step/phase of the user journey so it can optimized further.

Large-scale analytics techniques such as Process mining, Segmentation, and Clustering are employed here to identify deviations between expected and actual journeys and the time taken to progress through the steps and cluster the users into groups based on their intent. Identity resolution is another common data analytics technique that helps recognize the same user, and serve them personalized user experiences even when they switch devices and platforms. Deterministic identity resolution involves using first-party information such as login credentials to identify the users. Probabilistic identity resolution involves recognizing users based on non-primary user attributes such as their physical address or Wi-Fi location.

For example, YouTube identifies the same user logged-in user profiles on multiple devices using Deterministic Identity resolution methods. The same platform identifies users on the same Family plan based on their Wi-Fi location using Probabilistic identity stitching to make sure members not living in the same household on a regular basis cannot share a YouTube family account.

These Journey Analytics data points are then used to identify and store user preferences, which will be used during the Journey orchestration phase to deploy user experiences unique to individual user’s needs.

Journey Orchestration

Journey Orchestration is the process of visualizing, creating, and delivering optimized and personalized user experiences based on the learnings and observations delivered by the Journey mapping and Journey analysis exercises. The key differentiator that sets the Orchestration phase apart is the focus on personalization and the delivery of tailored User experiences for each Journey map and User Personas at a large scale. Journey Orchestration and hyper-personalization are made possible owing to the evolution of Machine learning models that can generate recommendations at a massive scale fueled by continuous learning based on data points gathered and analyzed during the Journey Analysis exercise. The insights and user preferences derived and stored from Journey analytics can also be used to predict user actions during this phase eventually resulting in a design that anticipates the user actions and aids them in reaching their goals quickly, hence delivering a friction-free user experience.

Conclusion

In a digital world that has millions of interaction opportunities competing for the User’s attention, having a strongly anchored focus on User Empathy goes a long way. Designers and Data experts can combine their knowledge base to create well-researched Journey Maps, Journey analyses, and Orchestrations that reduce cognitive burden and help users navigate Digital experiences in an accessible, usable, and delightful manner.

About the Author

Nithhyaa Ramamoorthy is a Data Subject matter Expert with over a decade’s worth of industry experience in Product analytics and Big Data, specifically in the intersection of healthcare and consumer behavior. She regularly contributes long-form thought leadership and career advice content to various Data Science publications. She is passionate about leveraging her analytics skills to drive business decisions that create inclusive and equitable digital products rooted in empathy. Opinions are her own.

References

• https://www.statista.com/statistics/617136/digital-population-worldwide/
• https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9998496/
• https://www.scientificamerican.com/article/why-we-are-wired-to-connect/