Pharma R&D Today
Ideas and Insight supporting all stages of Drug Discovery & Development
New Reaxys: A Model of Agile Development
Posted on April 11th, 2017 by Prima Sung in Chemistry
The development of a chemistry research solution is not something that should occur solely in the IT department of the product owner. Without the participation of the intended end users of the product, how can the developers be sure that they’re addressing real needs?
That’s why the development road map for the current version of Reaxys, Elsevier’s renowned chemistry research solution, allocated plenty of time for customer visits, ensuring that each idea would be discussed with the people whose opinions really matter.
To get a better picture of this customer-centric development, we met with Darryl Spencer, Vice President of Product Development, and Dr. Jürgen Swienty-Busch, Director of Product Management Chemistry. They shared the story of how New Reaxys was developed and where customers were involved along the way.
Reaxys was already a well-established research solution. Why did you feel it was necessary to upgrade it?
Jürgen: While it’s true that Reaxys is well-known and greatly appreciated for reaction and property data searches, when we talked to our customers, we noticed that they were actually ignoring a significant part of the content of Reaxys—and thus not necessarily getting the full benefit of it.
Many talked about using a general, not necessarily scientific search engine for the initial keyword-based searches to get an overview of a topic: a different dedicated solution for in-depth literature searches, and Reaxys for reaction and property searches. Reaxys had the potential to be used for all these search types but customers didn’t see that potential.
Darryl: Exactly. Looking at usage metrics, there was a clear gap between how Reaxys was being used and how it could be used. We wanted to make it easier for customers to get a greater percentage of their chemistry research needs covered by Reaxys—and that idea resonated with customers we talked to. Researchers are under an incredible amount of pressure, so getting more out of one tool would make their daily work easier.
Jürgen: Another concern was that the old user interface was not optimized for mobile devices and that is something that customers said was important. New Reaxys is designed so that you can use it anywhere you like.
What changes will researchers notice between the previous version and New Reaxys?
Darryl: Besides the obviously strikingly different user interface, there’s significant increase in the content covered, both in terms of journals and patents. We’ve added a vast amount of patents, including patents from offices in Asia—China, Japan, South Korea and Taiwan.
We’ve also expanded our indexing program. By indexing more, we’re making it easier for researchers to discover more. A chemist can now have much greater confidence that they’ll find all the information that’s available on a particular topic when they use Reaxys.
Jürgen: Increasing the discoverability of content also involved changing how people can search. Basically, users start with two types of queries: keyword-based and structure-based queries. Very rarely do they use the ability to enter complex advanced query types. This is reflected in the new simplified landing page: the Quick Search page allows them to find property data, reactions, literature, patents and experimental procedures by keyword searching, taxonomy-based phrase searching—including synonym searching—and structure-based searching. If there is still the need to enter complex queries, there is the new Query Builder, which helps each user to get more out of Reaxys. Even beginners can find their way around very quickly.
You both mentioned the difference in the user interface. What were the reasons for choosing the new design?
Jürgen: Users told us that they didn’t like the clutter in the previous user interface because they saw a lot of buttons but only used one or two. They found the rest distracting and unnecessary—it made Reaxys look more complicated and complex than it actually is. Therefore, we looked for a way to boil the user interface down to just the essential actions. We made it easier for chemists to find what they were looking for‑both in terms of actions in Reaxys and chemistry literature and data.
Darryl: The previous user interface was very much a product of the times when Reaxys first started. What we now consider the modern search engine was not a common thing that everyone used, so each research solution developer invented their interface and there was a tendency to put all the different functions up front. Now, everyone is familiar with the idea of a keyword-based search engine—everyone who’s been online has used one. It made sense to go in that direction.
Of course, we did both usability studies and design studies to verify our thinking! And it all pointed to the same thing: keep it simple, streamline it, show the essential functions.
New content is also something you’ve touched on. How was the new content chosen?
Darryl: We listened to users. The vast majority of Asian chemists and pharmacologists talked about the lack of patents from Asian offices, so we looked into that. We saw that not having that content meant that people would have to go to a different solution.
We’ll always respond to that kind of strong feedback about content, features and so on.
It certainly sounds like customer feedback was your main driver in this process.
Jürgen: Absolutely. Customer feedback grounds us. Everything we do should address a customer need. Talking to researchers means we understand what they need. Showing them what we are doing helps us see that we’re on the right track. We learn from them and we try to show them how we’re eliminating barriers to finding information.
Of course, we can’t make something that is perfect for everyone! But we certainly try to take on all feedback, consider all the possibilities. We have even incorporated a feedback button in Reaxys that allows end users to make suggestions, requests and comments whenever they need. It’s made it much easier to understand our customers—people use it, they appreciate the opportunity.
Darryl: We’ve discovered that we can innovate more rapidly by keeping customers involved all the time. Products like Reaxys should have agile development that involves end users in a responsive manner. You can’t sit idle for 3 years and then do a single beta test and expect to get enough information from it to make the necessary changes.
Jürgen: Beta tests are important, but they simply cannot give you the full picture. You get lots of feedback from a few beta testers and then others just do a couple of clicks and stop trying the product.
So we have interviews with customers all the time where customers can show us how they are using the product and how they’d like to use it. We have our feedback button that’s always available to end users. We run our big beta tests. And then we have focus groups, where we get into a room with 15 researchers for a full day, giving them ways to test every aspect of the product. That’s what we believe customer-driven development should involve.
Beyond that, we also have the Reaxys Advisory Board, which is a high-level group of researchers who consult with us about the product. We work very closely with usability experts, designers and psychologists to check our thinking all the way along. And whenever we encounter something that truly deserves more development, we adapt our plans in an agile manner.
What’s next for Reaxys? And what’s your biggest dream for this chemistry research solution?
Jürgen: Biggest dream? Right now, most researchers need 5 to 10 chemistry tools to do their work. I’d like them to get everything they need from one: Reaxys.
And so we are going to keep listening, keep looking at sources that might be important, keep finding ways to help people navigate it.
Darryl: I agree. We will continue to respond to feedback and keep improving search experience. We’re also looking at ways to help customers better integrate their content—their original data—with data they retrieve through Reaxys, so they have an even more streamlined workflow.
Because ultimately, we want researchers to have less challenges with data and more time for discovery and development.
Find out more about Reaxys.
All opinions shared in this post are the author’s own.
R&D Solutions for Pharma & Life SciencesWe're happy to discuss your needs and show you how Elsevier's Solution can help.
Sr. Product Marketing Manager at Elsevier
- Embracing the new in cheminformatics
- The Future of Chemistry R&D Will Depend on Deep Learning
- When Looking to Accelerate the Pace of Research, Having the Right Multidisciplinary Tools Matters
- To streamline early discovery, a cross-disciplinary, data-driven workflow process is key
- Chemists Concerned About How Their Field is Viewed