Mendeley Data :
The importance of Prototyping
Overview
Mendeley Data is a data repository that allows scientific researchers to publish their data in a way that can be referenced forever from a scientific paper. It is part of the Elsevier family and closely related to Mendeley - a world-class tool allowing writers to organise scientific papers and references. I also provided a way to search datasets from multiple organisations. I lead the design team in London to grow and improve Mendeley Data and look at ways to integrate it with other solutions in Elsevier, like Mendeley.
Problem
The repository tool was frequently used with lots of growth potential and did a small set of user cases well. We wanted to grow the user base and have more institutions feel the premium features were enough to pay for the tool. Part of that strategy was to turn the service into something researchers used as part of their research process.
Engaging
The London team was agile and engaged. It had created a valuable product and was eager to improve it. We also had a small team in Amsterdam, mainly management, with whom we regularly engaged. We also had an ongoing partnership with Manchester University, with regular visits to talk to Researchers, librarians, and IT staff. We built a working network of people involved with Mendeley data through regular catch-ups with design artefacts, one-to-one interviews, and structured group sessions.
During the project, Elsevier also bought other companies, such as an electronic Lab notebook, and we had other projects, such as data visualisation and the handling of scientific methods, added to the project. To manage all this added complexity, we required a north start prototype to start the conversation about our future delivery.
I held several group workshops at Institutions in which we worked on common problems researchers had
Internally we worked through our understanding of the steps researchers took
Discovery
We had a significant volume of analytics and an ongoing set of requests from the sales team. I joined and set up user research to fill in the gaps between our understanding of academic processes, the needs of those who bought the apps (staff in the institution) and the reality of the researchers we were engaged with. We usability tested the site to find minor issues, tapped into sales conversations, and initiated some structured workshops better to understand the everyday tasks and problems with researchers. We carried out workshops in Manchester and with institutions in France and Italy.
Through this discovery process, I created a detailed prototype that incorporated all the existing thinking and ideas and combined the new partnership tools. We showed it to people inside academia and got great feedback. The prototype’s most significant use was to make concrete all the different initiatives that were going on in the extended team.
It led to one significant realisation: given the current timeline, this was too big and complex to build. We iterated and simplified, but Mendeley Data could not quickly offer enough value to meet business targets. We needed to pivot.
We iterated through solutions using a complex functional prototype
Using the prototype we worked through difficult requirements such as linking to external data sources
The prototype grew from the existing solution and include a realistic version and north star version
Based on user research we worked out the architypal journey
Part of the discovery was working out how to unify the page templates we used
We also needed to work out the mental model of how different things were treated and who had ownership
Delivery
We balanced the rapid iterative design that included tactical changes with the ongoing strategic findings that fed into the overall Photobox approach. The roadmap we created was informed by the ongoing user research for prioritisation and the usability testing for designing new experiments to make different-sized tickets. In that way, the team could achieve a new constant velocity by having different-sized tasks. It also meant we had incremental changes going live that we could A/B test or monitor. We also used smaller markets to test releases before adding more larger markets. This allowed us to improve our certainty and reduce risk.
Many visual changes had negligible effects. I could demonstrate the power of UX by making minor changes in the App or Checkout flow, which led to measurable improvements in conversion. This was the advantage of working in an eCommerce environment.
The process could have been smoother. For example, when I returned from holiday at one point, I realised that the intended design/flow of one major part of the app had been changed to something that made more sense to the engineers. We worked out that despite clear documentation, the design needed to include more of the ‘why’ from the user research and involve more engineers. This change led to more engineer-led ideas and improved our process.
We evolved the data repository based upon our findings from our North Star and customer requests.
The dataset search tool was also updated to bring it inline with the Mendeley design system and unified.
I created a new structure for Mendeley data with templates to simplify development and ensure higher consistency.
Results
Engagement with the service grew over time, and we discovered that our ambition was more significant than the business could support, primarily by visualising this as a prototype. This meant we could rebalance the design team and have it focus on a more tactical direction. In this case, evolution was needed rather than revolution!
Prototyping leads to high levels of engagement and can expose issues with the company strategy, saving millions.
The tactical changes we found radically improved the satisfaction of users.
Learn
Prototypes are possible futures and allow businesses to better see the combined team aspirations.
Customers and users differ in many organisations; each group has different needs and perspectives.
Delivering against user needs is only sustainable if the business model works. These need to be balanced.
An isolated team can lead to miscommunication, and design teams need to maintain visibility and communication at the same level as any other part of the team.
Design Portfolio V5.9
Created using Framer