Data Health Dashboard
Data is one of the primary resources available for us to preform our work in the Office of Performance Management and Open Data. Both our open data program and performance measures depend on the reliability of the data we have. To help track the quality of our datasets we developed a few tools that allow us to monitor whats going on in our city data world.
Average Daily Views and Download
A key performance measure for the success of our program is tracking the average number of viewers both internal and external that visit the open data site as well as our internal data page. As we push for more communications with the public, better dashboard building, higher quality datasets needed by citizens, we want to see an increase in the number of downloads of our data and overall average daily views. After collecting some data, we should then get a better understanding of our baseline and will be able to set targets to measure ourselves against.
Job Log Table
Every time a dataset gets refreshed, a job log ends up here. If our workflow was successful the job returns as True. If there was a problem it returns False. This is great for a detailed look at how individual workflows are doing.
Job Log Timeline
Here's a more general visualization of how we're doing over time. You can track how many datasets we update every day and how many were successful versus failed. It might be helpful to see how things have gone over time, so feel free to adjust the filter in the top right of the visual. See the below graph for the same data grouped by month.
It's also important to keep track of what data is being utilized. These visualizations help us do that by tracking our most viewed and downloaded assets either all time or how those statistics are changing day to day.
Each night datasets are updated through workflows. An indicator that something might need attention is a workflow taking longer than expected to process. This visual is one way to keep an eye on our workflows and catch problems early.