Having worked on getting our data ready, we need to make sure we have an effective way to search it and discover research materials.

From week commencing 15 July, we’re bringing elements of the current Library search and the classic catalogue together into one search bar on the Library website homepage.

Over the past six months, we have been testing, researching, customising and tweaking to try to make the new search interface do everything we need it to do. We spoke to Katy Sidwell, our communications and web content manager, who worked as lead on this part of the project.

What did you need to consider ahead of launching the search project?

When we worked on the new website back in 2016, we did some testing and found that people had different ways of using the catalogue information – some people purely used Library search from the homepage and other people preferred the classic catalogue for some tasks. We knew from surveys carried out at the time that some staff really liked the classic catalogue too.

As we were moving from two different ways of searching to one combined search, we wanted to try to maintain the benefits from both search methods. There was no way of retaining the classic catalogue in the new system so we knew we would have to try to create something that was going to work for as many different scenarios as we could.

When sort of research and user testing did you do?

Before we actually got our hands on the new system in March, we did have access to a sandbox version using standard test data – it helped us to see how the system worked but not how it interpreted and presented different data types. It was more useful for us to search other institutions who have the same system so we reached out to a few universities and they were very gracious in answering our questions.

When it came time to do our user research, the first thing we wanted to do was understand how people search. We looked through the literature and back at previous studies we’d done when preparing the website. We also did some interviews to specifically try to understand the different contexts and approaches to searching and what people valued about the classic catalogue.

We then ran some observation sessions where we gave people tasks to do with the new and existing interfaces to see how people used each and adapted their search behaviour (or maybe they didn’t!) to find out what they actually did. We used this to determine the changes we made when we customised the new system.

What did you do with the research and testing data?

We had a big list of questions and requirements we gathered in the testing and research to work through when customising the interface. It was quite a big chunk of work to investigate what the system could and couldn’t deliver for us.

Research showed that a key task is known-item searching – especially on the staff side. People were coming to the catalogue knowing exactly what they wanted to find and wanted to see if we have it, where it is and how they can access it. For known-item specific searching, where you have the details, people valued the classic catalogue because it brought back fewer results. So we had to take that on board and try to give people the control over search results for known items so they appear closer to the top of the results.

What have you changed about the search to make it friendlier for fans of the classic catalogue?

What the classic catalogue had, and what we’ve tried to replicate on the new search, is some options. For those people who want a very specific item and know exactly what it is they’re looking for, they will be able to control the search by using the new drop down menus on the homepage. For example, if you want a book by Charlotte Brontë rather than a book or an article about her, you can choose the author option and the book option and that will produce a very focussed results set.

However, what we observed as the default search behaviour for everybody, regardless of system used, is you put in the title of the book or article into the search bar then hit return. So the options to refine your search results after performing a search are still there too.

The research was really useful. We tried to take as much into account as possible and get the interface as good as we can for when it goes live –but we’ll still have some work to do once we launch. Although we’ve done lots of testing, when more people start to use it we expect to hear about things that we haven’t previously accounted for. We’ll spend the first couple of months listening to the feedback and seeing what we can do to make iterative improvements.

What are some of the benefits of the new search interface?

For us behind the scenes, I really like the fact that we can change things in it – we were really restricted with the current search interface. It’s really good to have that flexibility so we can keep tinkering with it to try and make the search work for everybody’s needs. Unfortunately we can’t change all the things we’d like to!

In terms of using  the new search, there are loads of things you can do if you sign in before you search – you can save items for later and even save your searches. We could do this in the old system but they would disappear at the end of your session, which was no use if you were collecting references over a period of time. Not anymore! These will stay with your account until you’re done with them.

It brings together different editions of the same book – meaning you don’t get a long list with loads of different versions of a book. This makes it easier to scan results lists and see what editions are available. Similarly, if there is an ebook, this should be grouped with the physical book in the results.

There is a little feature that lets you see citations – you can see references in a journal article and references citing the article, and click through to them if they are also included in the catalogue. Before, that kind of functionality was only really available in some of our databases like Web of Science and SCOPUS.

Similarly, we’re working on the tool for citing references. In the item record, there’s a citation button which gives you the item reference and we’ve managed to link it up with the Leeds Harvard referencing style. Unfortunately we’re having trouble getting it to format correctly, which is kind of important in referencing! So this is something we’ll continue to work on after we go live. It would be great to offer one-click  copying of a reference into your bibliography.

One of the things we’re bringing in shortly after launch is one-click access for journal articles. The current user journey for accessing articles sometimes works well but sometimes isn’t great and needs a lot of clicks to get the download. This new service that we’re introducing cuts all of that out and presents a download PDF link in the brief results. We’re really excited about this! We’re hoping it will make researchers’ lives a lot easier.

You can explore what’s next on the Library website and view the other blogs in this series right here on the blog.