Getty's Museum Collection

In order to ensure that we understood the core motives that our researchers had when using Museum Collection (and how that might differ from our enthusiast audience), we decided to conduct a top task analysis. We collected a list of the user tasks defined in the product requirements document and added these to a simple survey on Optimal Workshop, asking 15 users to then arrange them in order of perceived importance.

The top five tasks for each group were the following:

The results from our survey showed us that the top three tasks for researchers were:

While these were also the same top three tasks for enthusiasts, an interesting difference was that the task voted on as the top priority by researchers was third for enthusiasts, emphasizing the researcher’s need for specificity in contrast to the enthusiast’s goal of more general discovery.

After determining our more general top tasks, we knew we needed to talk to actual users to get specific on our users’ behaviors and in particular the pain points that they encounter while navigating our application.

We decided to run two contextual inquiries – one with a staff member and another with a non-Getty museum professional, both of whom use the online collection on a regular basis. These sessions were conducted remotely to most closely mimic the actual work environment, but screen sharing allowed us to watch as our interviewees went along with their daily tasks, asking questions as needed. At the end of the two inquiries, we came away with a handful of key insights which are listed below.

After reviewing our research, we noticed that there were quite a few overlaps with the data that came out of both the contextual inquiries and user feedback. We took those insights and formulated two problem statements and corresponding design hypotheses that we felt we could turn around relatively quickly but would still make a positive impact on our researchers.

Problem Statement 1
‍
It is difficult for researchers to know which constituent is essential to them when multiple people and groups have the same or similar names.
‍
Design Hypothesis 1
‍
We believe that adding life dates to the search auto-complete will allow researchers to understand better which constituent is pertinent to their research question.

Problem Statement 2

‍Finding research-focused data is unintuitive.

Design Hypothesis 2
‍
We believe that placing the "History of the Artwork" section below "Full Record Details" will result in researchers intuitively finding the provenance information for an artwork faster.

Once we reviewed the pre-existing components and patterns in the Getty’s design system, we created multiple iterations of the design that we felt effectively solved the problem while still adhering to brand guidelines.

After a few rounds of internal review, we narrowed the options down to just three and then sent these designs to five Getty staff members – who were frequent users of Museum Collection – for a quick preference test.

This resulted in a split vote between two of the designs and some interesting feedback for both:

Vote for option 1:

“Dates are secondary information, so I think having them in grey will separate them from the different name options.”

Vote for option 2:
‍
“No. 2 isn’t as pretty as no. 1 but it’s faster to read, which is my top priority in collections stuff.”

Ultimately it seemed like having all the information on one line helped researchers scan faster, and as this was one of our primary user goals, we decided to move forward with this layout. We wanted to acknowledge, however, the aspects of the other designs that researchers liked and therefore decided to create a hybrid solution. We refined it further after one more round of feedback from the team and ended up with a design that helps researchers more easily find the information that they need.

The simplest solution was to flip these two sections so we created a prototype that reflected this. We then ran an unmoderated user test to better understand how users were navigating the page to access information (scrolling, selecting the jump link, etc.) and whether the layout change we had made mirrored their mental model.

Based on the feedback, we learned that flipping these two sections will likely help all users, but will be particularly helpful for those who use the jump link in the tombstone section as the typical behavior after jumping is to scroll down.

After implementing the change on the site, we heard via internal feedback that this was more closely aligned with how staff researchers expected the site to be organized and was a much welcome update.

This project is still in progress but as a next step it would be beneficial to run some user tests to ensure the design updates we’ve implemented thus far are successful with our core group of researchers. This would entail measuring how much faster users are able to find the search term they're looking for (in the case of the first design update) and testing whether the new page layout meets their expectations in terms of order of information (in the case of the second design update).

An interesting piece of feedback we received when testing the object page was that some users wanted a sort of sticky sidebar navigation to avoid the tediousness of scrolling up and down the page. We’ve started to explore design options for the existing sticky nav to incorporate jump links in the hopes that this will help provide researchers quick access to the information below the fold.