Springer Materials is an online resource for chemists, engineers and others involved in materials science, providing data and information on the physical properties of elements and compounds. As Senior UX Designer on the project, I have been responsible for developing the interface and experience on the site, ensuring that users can easily locate the information they need, designing interfaces for new features, and encouraging more sign-ups and customer retention.
Upon being brought on to the project, I was tasked with improving the site’s home page, as previous user research had proven that it provided a confusing experience to users. During testing, new visitors to the site often couldn’t tell what was on offer, or didn’t know how to reach the features they wanted to visit (such as they didn’t know that the two main images on the site were links to the structure and element search functions). Librarians, responsible for purchasing subscriptions to the site for their users in academic or corporate institutions, also felt that they weren’t able to see the full range of the databases that the site offered, especially the well known Landolt-Börnstein volumes, which the site was first based upon.
In order to solve this problem, we reworked the home page to show off the search functionalities more easily. We moved the links to the structure and element search functions close to the search bar so that it was clear to the user that they could search using more than just text. We also considered how best to display to a first-time user more clearly the various libraries of information, and provided a left-hand menu, which linked through to full listings on the search of everything in the specifically chosen library. Finally, in order to ensure that the full Landolt-Börnstein library could be more easily explored, we provided a separate link to a “bookshelf” feature, which listed each volume on a page, allowing the user to navigate down to the volume, then to the book, and finally to the page which they were interested in.
Finally, to help customer awareness of new features and additions to the site, we provided a “latest developments” section, where our marketing team could highlight specific sections of the site, providing information on the developments, and linking through from those stories to the specific pages, so the user could explore them at their leisure.
The improvements showed success from the day of release – we received feedback from users and librarians that they appreciated being able to see the full extent of the libraries on offer, and usage of both the structure and element search features increased after the changes had been made public.
Another issue which I discovered when analysing the structure of the site was the way in which the Information Architecture worked. In order to find a particular piece of information, the user had to take an unnecessary amount of steps to specify the information they required, and locate the actual information itself, which worried me would cause many users to leave the site in frustration, especially since the route required wasn’t conducive to the mental processes that the user expected to take. This problem was down to two main issues; firstly, the site had been built up in a modular fashion, with libraries of data being added on, and functionality added in such a fashion that the original user experience plan had been superseded by the sheer amount of information and “on demand” requirements of the development process, and also that some of the data had been provided in an archaic way – vital information, such as boiling points or chemical structures, were housed in PDF documents which the user had to locate, and then search through pages in order to find the required information.
In order to fully understand this problem, I provided further analysis by choosing key paths the user would take to find specific pieces of information, and illustrated those paths by devising user journey diagrams – representations of each page travelled through, indicating where clicks and interactions took place, and the number of steps required to complete the journey. These helped to explain the issues to key stakeholders, and suggest where we could provide improvements, as well as illustrate the intended processes for features we planned to implement in the future.
It’s worth noting here that these user journeys, as with the citation download user journey above, often went outside the pages of the site, and would involve referring to actions on PDFs, citation managers, or similar places where the journey would take the user. This is an important part of the user journey, as many have noted that UX extends beyond the screen, and the whole journey should be taken into account, to understand all of the implications involved.
As a result of the work I have done towards showing the user journeys, this has informed ongoing work towards digitising items that existed in PDF format, refining processes and pages so that users can specify what they want more quickly, and devising ways to use a knowledge graph search model to further improve contextual searching. I’m quite sure these benefits will continue to inform numerous decisions for the direction of the website in the future.
One of the most recent data sources to be added to the site is that of a Corrosion database. This resource provides information on the rate of corrosion of specific materials in specific environments, such as iron in seawater. Having been provided with the raw data, I was tasked with exploring how best to present this data to users, so that they could locate desired information more easily.
Firstly, I organised sessions with professionals in corrosion-related fields, in order to interview them and find out what information they required in their everyday work. Through this research, I found several important points:
- Whilst the basis of corrosion research consisted of a simple process (namely that a material plus an environment equals a rate of corrosion), there were other considerations, including coatings, which our corrosion feature had to take into consideration.
- Academic users cover a wider range of materials and applications
- Corporate users cover a smaller range of materials, focussing upon cost and lifespan of the material, with those in industries where calculating the rate of corrosion poorly could result in safety risks wanting very precise results.
- As with most of our other functions, the user journey ended up with an article, in order so that the user, either academic or corporate, could cite this source for their own research.
We built the function so that we could accommodate all of these requirements, and, if required, would allow for more precise and extensive information to be included later on. We ensured that the user could browse quickly through the information to find the information that they required and then allowed them to delve further into the specific set of data, get more detail and that all important reference to cite from.
The Springer Materials Corrosion function is now available on the website and has already received some good feedback from customers.
Previous incarnations of the Springer Materials website offered data on scanned PDF files, which provided a very complex method for users to obtain the data which they were looking for, as outlined in the user journey below:
In order to improve this process, I worked with Product Owners and stakeholders on the project to advocate that the data in the PDF files was digitised, so that it could be presented in a more accessible way. This took significant persuasion, as resources and money had to be devoted to scanning, editing, checking and verifying the data before it could be put into the system.
Thankfully, the company agreed, and therefore we were able to leverage the data to make it more presentable and usable. We were then able to design not only pages to present the data, but also methods of leveraging the data within the search to surface information to the user even before they got onto the page, as some of my sketches below indicate:
These were then developed into an interactive page which presented the data in a dynamic way, both in visual (graph) and data (table) format. The page also allows the user to change search details, manipulate the data set to specify certain ranges, as well as download the data for their own purposes, and cite the data set in their own papers.