It’s sad to close a service, especially one that you were involved in developing – but it is also a chance to reflect on what was learned from the process.
TREE was conceived in 2014, when our website was still using the somewhat clunky Polopoly content management system (CMS). The original vision was to develop an interactive tool aimed at teaching staff, to help them find out about the technology tools we offer and how these can be used for teaching and learning. We had ambitions to develop an interactive system that allowed people to rate our services add comments. We used extensive user consultation to develop requirements for this tool which I wrote about in a previous blog post – Setting out on an Agile project journey: Part 1, User Stories and Poker Chips.
TREE went live in summer of 2014. It was developed in Drupal – and its appearance was much cleaner than our webpages at the time. However, even with very limited content we had issues with the available search options.
Our main stumbling block however was with encouraging people to author content, and in keeping this content up to date. This required duplication of their effort as they were often already maintaining very similar content on service webpages.
We soon realised that there is too much overlap between TREE and our university webpages. Also, in 2015 our university website content management system changed from polopoly to Drupal. This has made it more user friendly to maintain webpages and offers better functionality than the previous CMS.
By creating a different web tool we were creating another place to direct staff too, this required an effort to promote, when staff were already going to our well recognised website in search of similar information.
We experimented with different channels for promotion – I set up the TREE Twitter account. I now see that I only posted 34 tweets – but I’m sure at least an hour of agony went into each composition! This was on top of email correspondence, print articles and some of my earliest blog posts. This experience introduced me to digital marketing and promotion and I learned that this requires a clear focus and vision and time to devote to development.
We learned from users – through comprehensive requirements gathering and later usability testing. People don’t all use websites in the way you think they will or that you do. Taking the time to consult with users on requirements and to sit with users during testing provides amazing insights.
We built relationships through requirements gathering and consultation and from this developed large mailing list.
What we learned:
If you can, put information where people are already looking for it – publicising a new website is hard! It’s much easier to improve content in the place people were going to look.
provide information that people can find for themselves
communicate the versatility of our tools.
Anyone can access the tool but it is primarily aimed at teaching staff, particularly those new to the University.
I wrote about this project in a previous post where I talked about the requirements gathering process. To recap- we used an agile methodology for this project, keeping users at the centre of development throughout, from user stories for gathering requirements to frequent user testing.
I can’t believe how far we have come in the six months since I wrote that post. At that point it was still called the ‘Resource Discovery Tool’.
What’s in a name?
We started thinking about what to call this tool early in the project. ‘Resource Discovery Tool’ was too broad and non-specific, and for librarians means something more akin to a library catalogue. But it did lead us to some interesting conversations with our library colleagues. We collected name suggestions from participants in the requirements gathering sessions and then set up a survey in Bristol Online Surveys to poll opinion on these suggestions:
ASK (Academic Services Knowledgebase)
ASK FRED (Academic Services Knowledgebase Finding Resource Education Discovery)
TREE (Technology Resources for Educational Enhancement)
And the winner was: ASK (Academic Services Knowledgebase) (with ASK FRED a close second).
So why didn’t we call it that? Well, one of the fun things about working in a large institution is that there is so much going on, and it turns out that this name is already earmarked for another project. So we went back to the drawing board and sent out a second survey offering the choices:
GUIDE (Guide to University Information supporting Digital Education)
TREE (Technology Resources for Educational Enhancement)
And so the winner was TREE (Technology Resources for Educational Enhancement). We were pleased with this name as we hope this name reflects what the tool is for and the acronym TREE offers interesting visual options for branding. Alas, it also offers Steph a world of pun options and word play became a large part of the TREE meetings and quickly infected our communications!
Growing TREE, from seed to sapling
We decided to build in Drupal, as the university website is also moving to this and there is in-house expertise to call upon when advice was needed. Continuing the Agile approach we went through four short build and checking cycles to reach this first live version. I say we, but really it was Steph who has worked the magic in Drupal on this project.
For each build we also undertook user acceptance testing (UAT), to check TREE was functioning as planned. For the first two rounds we send out test scripts (with a list of tasks) and asked users to complete these and email them back to us. This was great because testers could complete these when it suited them, from their own desks. However, it was difficult to frame tasks to test all aspects of the tool (for example how people react to a null search response) and it was not always clear from notes what testers had done at points. While the UAT was useful and successful, it did not allow us to do as much usability testing as we would have liked.
So for the last two rounds of testing we tried another approach called ‘accompanied surfing’. For this we booked a room and invited users in to sit with us as they worked through the tasks on the test script, talking through their thought processes as they did so. This offered a far greater insight into users expectations and search strategies than the earlier User Acceptance Testing.
I’m really pleased to have been involved in this project and it’s really satisfying for Steph and I to have created a live web tool, it’s a great feeling to see those user requirements turn into a real functional tool.
I’m also really pleased that this is just the start for TREE, as a phase two project has been agreed. Future developments may include:
extending the range of services included beyond those offered by IS
adding functionality to rate or comment on services.
We hope that TREE will be the ‘go-to’ site for any staff member at the University of Edinburgh interested in central services for technology enhanced learning.
Seeing the TREE in the woods?
Alongside further development comes raising awareness. We want to tell everyone involved in teaching in the University of Edinburgh about the TREE tool.
We are fortunate that so many people at the university contributed to this project and would like to thank all of them:
17 participated in requirements gathering workshops
143 completed the requirements gathering survey
14 took part in prioritisation events
47 individuals took part in user testing
many more provided help and consultation.
This is a large institution and it will take effort to make sure all staff in a teaching role are aware of the TREE tool. We are sending out emails and news items to every list we can access but would really appreciate help with this.
So please tell people about TREE
Please can you forward the TREE link (http://www.tree.is.ed.ac.uk/) to any list or person you think might be interested. Please add the link to any webpages, wiki pages or documents where you think it would be relevant, for example those you give to new members of staff.
I look forward to letting you know how the next phase of development goes in a future post.