The SF Department of Public Works is in charge of all roadworks in the city, as well as any construction at libraries, the airport, and a number of other large projects. When a contractor bids on a job, their information is compared to historical data from past bids.
Currently, this information is kept on many spreadsheets across many computers across many departments within the DPW. All of this information was being manually entered into a central SQL database, and I was brought in to design the front end interface for it.
I focused on bringing it up to interaction standards, and ensuring it fit within the larger DPW branding.
After presenting paper prototypes, I had a rudimentary interface to some "dummy data" available for user testing. I convinced the stakeholders that user testing was actually important, and eventually managed a week of conversations with people who would be interacting with the database in a variety of ways.
Unfortunately, while I was in the process of pulling together a report of the user feedback, the project was canceled.
I was the only person working on design / planning / research / UX for this project. I also started building a branding guidelines document, since the original could not be provided.
The biggest challenge came towards the end of the project, after I had been able to speak to representatives of targeted user groups. The stakeholders who provided much of the information were only one of several of these groups, and had made no attempt to understand the other ways these data were being used.
emacs, Inkscape, Gimp, KendoUI, highlighter and paper.
After producing some very well-received paper prototypes to ensure that I was on the right track, my next task was to convince the stakeholders that user research was necessary for the next step. While those gears turned, I worked on HTML and CSS files left by the previous designer, and was able to use those as a good start for both the front end and the branding. I was then told to switch to a webapp using the KendoUI framework.
Once I had the new prototype hooked up to a set of "dummy data," I had been given access to users who worked at all levels of the interaction chain (input, search, report generation, etc...) for this dataset. A week of interviews generated a huge amount of data that indicated a much wider scope of project.
While I was elbow-deep in sorting and parsing and coallating this information, the project was canceled.