To examine how user-friendly our website was, we conducted some user studies on the first version of our website. We had approximately 20 participants who we invited to try out the website. After they spent some time exploring the features, we asked them to fill out a survey. Important questions from the survey included:

• What was your first impression on the website?

• Do you have any feedback on the UI?

• Are there any additional features you wish to see?

• How easy was it to navigate the website?

• Would you use this website in the future if you need datasets about climate change?

• How would you rate the user interface?

• Anything else you'd like to share with us?

The survey received positive feedback from users, with 100% indicating that they would check NADBenchmark if they need datasets about climate change, 86% of users find the website easy to navigate, and 92% like the UI.

We also received insights on what could be done to make the website experience better, and make improvement changes accordingly.

Improvement Suggestions

Data submission form

The dataset submission form in the first website version was long, and some feedback we received mentioned that the form seems to be intimidatingly long because users needed to scroll down continuously to fill out all the information. In the current version, we break the form into smaller pages so that users can fill in small chunks of information and therefore are less impatient when filling out the form.

Make the Submit Data a separate navigation bar category

Initially, users needed to click "Contact", from which they can see the options to either "Email us", or "Submit dataset". Several users said that we should make the "Submit dataset" option more verbose and easy to find. Therefore, in the latest website version, we decided to separate this feature from the "Contact" page and make it a separate category in the navigation bar.

No suggestion for dataset search bar

Before the user evaluation, we believed that users would love to have cues while searching for datasets, hence built our prototype with a lot of suggestion features. However, the feedback showed that users find these suggestions unnecessary. At the improvement stage, we get rid of the suggestion feature and leave more space for the search bar