Taboola platform allows online content publishers to market digital content through a recommendation engine which is served via site widgets and feeds on publishers and websites across the web.
One of the key features of Newsroom is alerting users when a notable event is occurring on a publisher's article. Newsroom has alerts for important events including Overview, Front Page, and Network Insights. The alerts include many different scenarios, providing both positive and negative feedback, including traffic increases/drop-offs, instant hits, old articles with new traffic, and several alerts for social media traffic including Facebook and Reddit.
Publishers and staff have no way to turn alerts off outside their work shifts or vacation without turning off alerts globally — an impractical solution. News happens 24/7, and some publishers have a staff that works all hours of the day and weekends. Publisher staffers could work a consistent weekly schedule with shifts that could be any time of the day, or differ day-to-day. A shift could start in one day and end in another day.
The goal was to design a more nuanced way to allow alerts to come through to the staffers at a time they would be able to see and act on them. The design solution had to be flexible enough to account for the aforementioned edge cases around staffers' work shifts. During off-hours, the alerts would queue up and display when a work shift begins. We also needed to account for a temporary absence, such as a vacation, that would override the set time windows until the conclusion of the vacation.
As the sole product designer, my role was to research all the possible scenarios the user experience would need to account for and come up with a flexible design solution that would work for consistent schedules and varied schedules.
To understand more about whom I was designing for, I started by conducting interviews with publisher staff and our internal team, some of who had experience in a large newsroom. The goal was to learn more about work shifts/schedules, publisher roles, and absences from the publisher newsroom.
After considering all the use cases, it was decided to give the publisher staff two paths to set up their time windows. The first path, and the most simplistic, was for staffers with a consistent schedule. Generally speaking, this would be for the 8-5 weekday shifts.
The second path was for the user who had a more complicated schedule. I called this the varied alert window. This path covers all the edge cases that the first path misses, such as staffers who have an inconsistent weekly schedule, a part-time staffer, or a staffer who works "graveyard" shift that runs past midnight into the next day. An editorial staffer could add time windows by clicking on the start hour, then clicking on the end hour (We decided to use one-hour increments for simplification).
When a staffer clicks on the start hour, it starts the selection task with an arrow pointed forward highlighting all the hours between the start time and the position of the cursor. When the user clicks an end hour in the alert window, the user interaction is complete, with all the hours in the window highlighted. This establishes the alert window.
To remove the alert window, the user can hover over the end hour and a delete button would appear. Above the weekly calendar, the weekly alert schedule is in natural language for clarity.
Also included in this feature is vacation mode, this allows anybody on the editorial staff to set a one-off period in the future where alerts are muted and queued. Vacation mode would override the established alert windows until the vacation ends. If vacation mode is only one day, then the end day is optional.
For the varied alert window, there was one small edge case that we found this design did not cover well. When a night shift ran from Sunday evening through Monday morning, the user would have to click on the first date and then move the cursor a long distance back up to Sunday. As a workaround, the user can create an alert window from Sunday evening until midnight and then create another alert window Monday morning until the end of the work shift. It is a small amount of additional work for the user, but we decided it was an edge case not worth solving at this time.
It is also worth noting that there is already a user experience paradigm for blocking out times in a day, calendar apps from Apple and Google already use the design familiar to most people who spend time working in an office. In my opinion, this would have provided a better user experience because it would have provided a visual queue for blocked out times that would have been more familiar to our users. However, it was not technically feasible within the scope alotted for this feature.