Introducing — Systems UX Research
For improved human-computer interoperability
The Systems User Experience (UX) Researcher is responsible for articulating how and in what way the perceptions, sayings, and doings of people are ingested by and accurately represented in the broader network of technical systems (e.g. as data, input, conditions, etc.).
In this article, I will be introducing a new branch of User Experience (UX) Research known as “Systems UX Research.” It builds from the fundamentals of Human Computer Interaction (HCI) and UX, to logically extend the present focus beyond user interactions with interfaces, to attend to the integration and representation of user generated input in back-end coded systems. With the extension of UX Research’s purview into technical systems, comes a need for the UX researcher to be conversationally fluent in and practically skilled in the makings and workings of technical systems. This is paramount for the identification of representational or integration issues of user generated data and signals into computational systems and the successful advocacy for their improvement.
Acknowledging and articulating the method and practice of Systems UX Research is increasingly relevant with the rapid growth of new AI applications in consumer products and features.(For those unfamiliar with AI, GenAI, and LLMs, please see brief informational note at the end of article). AI technology draws from a manifold of user-derived data points and signals to deliver targeted ads, “relevant” query results, personalized experiences, and content creation. The feedback-loop between human-and-computers with AI is accelerated, large volumes of user data (specific and general) is gathered and analyzed seemingly real-time, allowing for increased interactional frequency between a user (input) and AI (output). The representation of users within high-touch technical systems, be that as machine readable language or data points, is more prone to errors in equal proportion to the impact a misrepresentation or misinterpretation can have on the experiential quality (and business liquidity) of said product or service.
UX professionals as Systems UXers will need to be able to mind the current gap forming between a person’s interaction with a highly complex technical product and the follow on translation of said user activity and information into computer readable language or data as it gets integrated into programmed computer applications. No one else within the tech industry is attending to this growing gap, rather the increase in specialization and siloeing across engineers, product managers, program managers, UX, marketing, and data analysts has exacerbated this disconnect. It is time we consider user-centered systems thinking with the establishment of Systems UXR.
What are Systems?
At this point, you might be hung up on the word “systems”, so let me take a moment to unpack the working definition for “systems” here and then we can proceed with its relevance to UX Research as Systems UX Research.
In software engineering, “systems” is a commonly used term to refer to operating systems (for example MacOS, Linux, ChromeOS, etc.) and their architectures. From a technology standpoint, systems are a compilation of hardware and software that are made to work together to achieve a certain objective or goal. These pieces of technology are often developed at different times, by different people, even different companies, and require additional supporting software, like Application Programming Interfaces (APIs), to serve as translators between otherwise foreign systems to allow for inter-communicability. The study of this relation between otherwise disparate systems to ensure successful and predictable interoperability is the work of Systems Research (primary source: Wikipedia, “Systems.” accessed Aug. 10, 2023).
A Systems Researcher is interested in understanding how systems work, how multiple sub-systems can be brought together towards realizing common objectives or ends, and how their performance can be optimized with increased cross-system interoperability. Comparatively, Systems UX Research is interested in understanding how humans and computers come together to form a larger and ideally interoperable system that realizes aligned objectives.
If we make a small conceptual leap and regard people (i.e. users of technical products and services) as a type of “system” not unlike operating systems, then we can easily pivot to a discussion of Systems UX Research as focused on ensuring the smooth translation between two otherwise foreign systems — people-systems and technical systems — to ensure inter-communicability. The role of a Systems UXR is to ensure there are support layers in place at all human-technical systems touch points for successful cooperation, or interoperability. This is a vital part of ensuring the overall efficiency and performance of complex human-technical systems.
Systems UX Research
A Systems UXR attends to the communicability of people-as-systems with technical systems and vice-versa, between technical systems and people-as-systems, to ensure there are no errors introduced in the beginning, middle, or end of the larger human-technical system that forms. I’d argue that this is a novel approach to UX; regarding people-as-systems as continuous and contiguous with technical systems, requiring a more holistic lens in our treatment thereof.
Presently the profession of User Experience finds itself confined in its discipline and practice to the realm of people-as-systems and user interfaces, excluded from expanding a user-centered lens to the technical systems beyond these touch points. While said touch points between human-and-technical systems are of extreme importance, limiting UX purview to said micro-interactions does not address macro-system level issues that can easily undercut any gains in human-technical system interoperability won through user interface improvements.
Hitherto, little time and attention has been given to that broader integration and communicability of people-as-systems with the complex of interoperable technical systems. Within the tech sector, it is the User Experience Professional that is charged with understanding and advocating for the inclusion of user-centered thinking in product ideation, design, and delivery. It is also, as I’ve argued here, the User Experience Professional that is best suited for identifying and advocating for improved representation, translation, and processual handling of user-derived data and user-generated signals in the technical systems it is integrated into and passes out of as output.
The systems one must address as “Systems UX Researcher” are veritable concatenations of subsystems with multiple translation layers (including micro-services with manifold API layers, coding language changes, and decade old legacy systems). Each system touch point, be that between human and technical system or technical system with technical system, is a point of risk wherein user signals or data may be improperly handled, mutated, and wholly abandoned.
As a Systems UXR you will need to understand the full feedback-loop of user generated input throughout the broader complex of systems to ensure it is properly processed, with minimal degeneration or mutation. This requires an appreciable, but not expert, level of technical comprehension to allow for mapping end-to-end the journey of user data and input through technical systems. To this end, numerous IT companies are encouraging their UX professionals to invest in learning basics of programming and data analytics, even offering various continued education e-courses at nominal fees.
Systems UX Research in Practice
We should never expect our users to understand the “black box” of technicity behind the glossy user interface, but that doesn’t mean we as UX professionals shouldn’t attend to it (see my previous article on user experience black boxes). A user’s experience with any technology, their inputs and it’s outputs should be intuitive, expected, and consistent because the end-to-end system processing of their input should not be introducing erroneous mutations or changes. Inconsistencies or a lack of intuitiveness in a user’s experience are signals that there are one or more systems’ interoperability issues that need to be explored and resolved. It may very well be that the issue is with the user interface, the touch point between people-as-systems and technical system, but a Systems UX professional’s attention should not be confined to that interface, rather we should be mapping the entirety of human-and-computer systems to identify all possible causes across the broader feedback loop:
Mapping Human-Computer Systems
- Identify all forms of existing user-input, including content/information, state changes, settings, or engagement, etc.
- Control for the user interface to ensure there aren’t any usability related issues lending to interoperability issues
- Identify all systems’ output that leverage user-input and correlate input → output/results
- Note all points of translation or mutation in user generated signals throughout its processing in systems.
Adding one more note of caution, our technical systems should remain black-boxes — invisible, simple, and predictable — from the users’ point of view. It is not their responsibility to understand signal mutation and representational values of data across various technical programs and applications so as to “comprehend” the output. What they put in and what comes out should “hold together” as consistent and coherent sides of the same coin. It is our job as Systems UX professionals to ensure the submersed portion of the systems’ iceberg is sound.
What’s Next?
Upon reading this article, many a UX professional and researcher will discover that they already perform aspects of Systems UX Research. In part because it is a logical extension of many of the core principles and functions of HCI and UX Research. Regardless of whether you have tried your hand in Systems UX Research, I hope that this articulation is a catalyst for further discourse on and around the method and practice of what comes to be more broadly recognized as Systems UX Research.
Quick note on terminology:
- Artificial Intelligence (AI) is a broad class of machine learning applications that automate large-scale data analysis or computational programs for the purpose of executing certain functions such as targeted advertisement, user personalization, directional guidance, data queries and much more.
- Generative AI (GenAI) is the application of artificial intelligence for generating content. The most publicly recognized form of this includes Microsoft & OpenAI’s ChatGPT, but can also include image generation, video generation, and code generation.
- Large Language Models (LLMs) are also AI that, as the name implies, can take very large and complex data sets of linguistic content and develop representational models of that language. These models can be applied to generate content (as in the case of ChatGPT and Bard) or perform computations.
