Our research progressed into the development of four core scenarios that engage data tracks.
The term Data Tracks describes paths of information generated by a user in order to measure, manage, and explore themselves through their data. Data tracks differ from data collected about a user for various reasons. Data tracks are generated by a motivation to receive personal benefit, thus they embed an honest, first-person perception. Unlike observed data, which may be saturated with observer bias, data tracks are interwoven with the complexity of life, such that they are unfiltered sources of information. Finally, data tracks are typically time-based and can reveal change and transformation. This dynamic dimension lends itself to greater holistic trends and analytical analysis.
Organizations that seek to be human-centered and data-driven may not consider the interaction between the two. In the following frameworks our team explores how personal, scalable, and welcomed data products are developed in collaboration with data tracks. These frameworks are tools for thinking which design, research, and technology teams can discuss to inspire human-centered data-driven (HCDD) solutions.
Locating Data in Experience
A Human-Centered Data Driven (HCDD) Competitive Landscape
Technology companies may not all be designed for user experiences but they can all be informed by them. Whether or not the product, service, or solution reaches an end-user or an intermediary, it is important to understand the layers in which these data-oriented services are located.
The competitive landscape for data products ranges from wearable devices, entertainment algorithms, DNA heritage, and analytical cloud services. Despite the diversity of these applications, they can be filtered through four areas of user experience: Intent, Affordance, Context, and Analysis.
Filters of Experience
- Intent: Self data, originating from user directed activities
- Affordance: Data which arrives based on possible actions or behaviors from the user
- Context: Data produced adjacent to user activities and environments or afforded technology
- Analysis: Data resulting from gathered data sources relating to the user indirectly
Is your organization vertically integrated for data?
As an analog to vertically integrated supply chains, in which expansion is achieved through direct ownership of various stages of its production process, enterprises can be vertically integrated for data. Google, Apple, Netflix, and Meta, among others, are valued as data companies over product or service companies. This is because they have installed waypoints along the terrain of user experience. They not only build a product but can create those products with their self-created data backbone.
Is your organization horizontally integrated for data?
Likewise, the horizontal integration strategy holds benefits through increasing share of production at the same node of a supply chain. Horizontal integration for data is seen through companies like Salesforce, AWS, SAP, and Oracle, which provide data and digital services platforms for various clients. These organizations sit at the bottom of the user experience terrain, and are reliant on the data sources which cascade from above.
There may not be a “right” data strategy, as both may result in success. But the recognition that data begins at user intent and transforms itself on the way down is critical to maintaining an advantage in the HCDD competitive landscape.
Designing with Time
Extending Data Lifecycle
Acknowledging users change through time, and designing systems that adapt with change.
Needs and jobs change throughout the life of users, businesses, and communities. Most existing forms of self tracking focus on a specific concern bounded by time, such as products segmented by life stage, like baby monitors. Some target discrete goals but are difficult to find value from outside of those goals, as with weight management apps. Others are generalized across multiple experiences, but are often disconnected from user identity, as with smartwatches that are generalized for adult fitness.
Systems that are cognizant of the passage of time are able to use data tracks more effectively. These solutions bring focus to transition and can adapt to long term change. As users age from adolescent to geriatric, so do civic institutions from municipalities to nations, and organizations from start-ups to multinational corporations.
Strategies which design for change embed an awareness of the possibilities and limitations of data capture. As people and organizations age, considering how a user experience and its data change throughout a user lifecycle is a significant step to advancing innovation.
Human-Centered Data Governance
Managing the Flow of Information
Understanding the holistic value and responsibility of managing information flow from data generator to data custodian.
Data-driven systems have clear value for agile organizations. But without a clear understanding of the data generating sources, the role of the data custodian, and the filters that lie between, organizations lack holistic impact.
We have developed new terminology for the following HCDD roles:
- Data Generator: an information source where data is generated passively, actively, automatically, or manually. May include the human body, the environment, and other measurable process outputs.
- Data Custodian: an entity which governs the flow of information, manages data processes, and protects the privacy of data generators.
In the process of collecting data, the transformations mimic system loss, not unlike the game of telephone. Users generate data but only some sources are allowed to flow by way of affordances. For example, only so much information about one’s mood can be gathered on a wrist via a smartwatch. As the information travels, other forms of information join in to balance data scarcity, such as synthetic data, historical data, and anonymized data. The context layer may limit information due to social norms, culture, and lived experience. Lastly but with the most complexity, data processing due to analysis may use algorithms to draw the signal from the noise. From an engineering perspective, much is simplified when signals are clear for decision making. However, noisy data is not necessarily invaluable data; oversimplification may lead to unintended outcomes.
Human-centered design methods focus on the end-user in order to involve them and their experience in the development process. Data-driven organizations can benefit from this methodology by focusing on individual user data to benefit collective understanding, unlike traditional customer segmentation which forces users to conform to a persona. Data governance is required to ensure the flow of information is robust, ethical, and impactful.
Involving the Self
Designing for Situational Self-Efficacy
Actively engaging users creates the opportunity for introspection and change.
As the gap between digital and virtual diminishes, as a culture, we have seen an increasing need for unplugging and reconnecting with the self.
Most self-tracking technologies which focus on improving self-efficacy rely on passively collecting, analyzing, and applying information collected from biomarkers. More so, the data collected is personal to an individual, yet minimal steps are taken to prevent data misuse. Considering user vulnerability, regardless of the distance of the user to the product, is paramount to creating value in a fearful culture.
In such a situation, it becomes critical to focus on actively involving the self in generation, analysis, and interpretation of personal data, which could be done via the individual building their own data literacy. Since HCDD deals with personal data, it’s imperative to take ethically responsible decisions to secure sensitive information. This could be done via providing ownership of data to the individual and decentralization of data sources, such as using technologies like edge computing or blockchain.
Ecologies of Scale
Supporting Self-Tracking Communities
Data generated by individuals can inform adjacent collective layers through ethical systems.
Collecting individual data from the bottom-up helps organizations improve diversity and inclusiveness and broaden the population of products, services, and systems; while top-down approaches in the market mainly target normalized or mainstream people and may ignore marginalized voices.
In encouraging users to share their data tracks, user agreement on data sharing is a key to scale up HCDD solutions. Designing persuasive, accessible, and user-friendly experiences is important to nudge people to share their data both for individual and collective benefit.
Organizations must set boundaries to improve data privacy, data accessibility, and data ownership in order to build trustworthy and well-balanced systems that can know the value as well as the risks of collective data. For example, a subset of users (individuals, families) might access private data while others are limited to filtered data to mitigate risks of data leakage. Regarding data ownership, a system could allow individuals to delete or revoke their data across the system. Similarly, community members could agree on a requirement to delete data valuable for the community. Avoiding polarized attitudes on those sensitive issues gives an opportunity to achieve system optimization without exploiting individual humans.
Collective data has a potential to provide value where individual data cannot. In a smart city context, city governments may build customizable infrastructures in which service providers can offer consistent services on top of infrastructure for specific user groups throughout their lifetime. On the other hand, newly generated risks should be taken into consideration for different time frames, user groups, and data applications.
In designing sustainable and scalable HCDD systems, questions to ask are:
- What could be the benefits of collective data to different system layers in the long term?
- What could be unexpected consequences or risks in the long term? What is being done to prevent them?
- How should data be stored for in the short term versus the long term? Can it and how might it be encrypted to minimize risks?
Our team believes the self is under threat. As the wave of unplugging, data breaches, and privacy infringement concerns haunt our present culture, numerous wearable makers remain unaware of the needs of the user and, more importantly, the wellbeing of their data tracks.
The strategies above surfaced from the Memiro prototyping process, which centers on four unique cases of user vulnerability. In the following blog posts we will detail how each of the scenarios exemplifies the value of data tracks at the user level, and additionally as they scale to enterprise, civic, or institutional levels.
Memiro — The Future of Data Tracks
- Synthesizing Memory
- Finding Alignment through Transition
- Encrypting Understanding
- Gesturing Community Safety
