Many organizations control vast amounts of data but struggle to unlock its value. A promising approach is to democratize access to data, maximizing the number of people in the organization that can benefit from it. However, the importance of the end user experience (UX) is often overlooked in that regard. An IT platform that enables access to data is of little value if people fail to embrace and successfully use that platform. Here, we discuss how ONTOFORCE addresses this challenge.
The UX of self-service data access
In the continuous quest to unlock the value of data, a hot topic is self-service data access: allowing as many people as possible in the organization to explore data and put it to use.
In order to ensure that people actually embrace such a self-service data access and exploration platform, it’s important to first understand what people expect from it. And when it comes to extracting value from data, there is no one-size-fits-all approach. At ONTOFORCE, we categorize the diverse expectations of the user audience according to two main dimensions:
(1) DIFFERENT LEVELS OF EXPERTISE
The first dimension captures the IT skills and interests of a user:
(2) SEARCHING FOR THE KNOWN OR SEARCHING FOR THE UNKNOWN
A second crucial – but often overlooked – dimension is how a user expects to interact with the data platform:
Combining both dimensions provides a basic but valuable framework to help understand the potential breadth and depth of end user expectations. For example, the two extreme cells in this matrix indicate completely different end user expectations:
DISQOVER is an advanced knowledge discovery platform that leverages powerful data science concepts such as linked data, knowledge graphs and semantic search. With this platform, we rely on three pillars to address the wide range of expectations concerning end user experience: interactivity, stratified complexity and customizability.
With the DISQOVER platform, a user always gets results in a matter of seconds. Combined with an attractive and engaging user interface, users are encouraged to further explore and drill deeper into the data. This potentially results in the discovery of information nuggets and valuable insights that would have remained hidden to the user had they ended the exploration journey earlier – for example, on a less engaging platform where every step takes a long time to complete.
Software tools usually fall under one of the following categories:
A mismatch in expectations often leads to poor user adaptation: users looking for a simple solution might be scared off by the complexity of an expert tool, and users looking for power and flexibility might be disappointed by the lack of it in a general purpose tool.
Our strategy with DISQOVER is to present it as a simple, general purpose tool to a novice user, with the ability to proceed to an advanced layer with expert features once a user embraces it and appreciates its value. In this way, we avoid users being disappointed due to a mismatch in expectations, and we strive to get the best of both worlds when it comes to productivity over time.
DISQOVER can be deeply customized to match the user’s expectations concerning dashboards, reports and navigational journeys, even to the level of adapting the terminology used in the product to specific use cases.
Data scientists can prepare a completely customized and simplified user experience that is tuned to the key use cases of the employees of an organization. Importantly, these customizations are made via the product’s graphical user interface using drag-and-drop and step-by-step wizards. There is no need to learn to write complex scripts or toggle configuration files. This introduces the concept of self-service customizations: expert users in a team can easily create customized views and roll these out to their peers.
With the upcoming 6.00 version of DISQOVER, we take this concept an important step further by allowing the user experience to be fully customized from the very first step in a discovery journey: the search page.
Addressing all expectations
DISQOVER’s UX strategy makes it possible to address all aspects of the end user expectation matrix and empowers an individual to progress through platform functionalities in a natural way. A new user typically starts by executing preconfigured use cases and can proceed to expert exploratory knowledge discovery over time.
An example: finding expertises
An interesting use case that can be addressed with the DISQOVER platform is finding people with a specific expertise. For example, suppose you search for key opinion leaders regarding the treatment of a certain disease by posing the question, “Who is leading active clinical trials concerning this disease?”
STEP 1: AN EXPERT USER FINDS THE ANSWER IN INTERACTIVE DISCOVERY MODE
An expert user answers that question using DISQOVER in an interactive, exploratory way by creating the following navigational path, leveraging the advanced linked data capabilities of the platform:
The near real-time responsiveness of DISQOVER ensures that this navigational journey takes less than a minute to complete, assuming that the user is skilled and clearly understands the platform’s linked data concepts and the underlying data model.
The DISQOVER search page in expert/unscripted discovery mode, providing direct access to all available data. This is a powerful view, but it is potentially overwhelming for non-expert users.
STEP 2: ROLLING THIS OUT AS A PREPARED USE CASE TO ALL USERS
With version 6.00 in the above use case, the expert user can subsequently customize the start page of DISQOVER and add a simple button that automatically applies this full navigational journey to any disease term entered by the user. From now on, every user is capable of easily and correctly answering this question for any disease by simply clicking a single button, within a couple of seconds, and without requiring any specific background knowledge of the platform and the underlying data model.
A customized DISQOVER search page for expertise finding with three possible approaches prepared that are accessible via simple buttons and include clear explanations. Note that a button leading to the advanced view is still present on this search page, allowing the user to easily switch to the more advanced, unscripted data exploration mode.
Summing it up
Democratizing access to data is a promising approach to help unlock the value of data, but even the most advanced technology is of little value if people do not embrace it. This is a lesson that many businesses have learned the hard way because they focused too much on the technology, with user friendliness and user experience only being an afterthought.
In order to avoid this pitfall, it is paramount to properly understand end user expectations and build the platform from the ground up with a high quality, personalised user experience in mind. Most of all, it is key to acknowledge that the intended audience, even within a single organization, can be very diverse and must be properly segmented, and that the platform’s user experience should be diversified in order to match the specific needs of each user segment.
With DISQOVER, we address this challenge by building a deeply customizable platform, making it possible to offer each type of user a personalized user experience. This ranges from answering prepared use cases with the click of a button to enabling advanced, interactive and unscripted knowledge discovery.