“Faced with the choice between changing one’s mind and proving that there is no need to do so, almost everyone gets busy on the proof.” – John Kenneth Galbraith
Within the world of open data, the concepts of intrinsic value for the greater good and proving value through hard analytics exist in an often uncomfortable partnership. There are good arguments on both sides, and when data can back up the intrinsic good, it can be a win-win for everyone invested in the outcome. When, on the other hand, data contradicts the assumptions made within the initial idea, that can be of value as well.
Here is one example of how APPCityLife’s real time mobile analytics platform is helping our civic clients and our own development team to manage, verify, and justify open data mobile initiatives.
Open Data Mobile App: Verification of Value
Preparing a set of data so that it is not only accessible but useful to the public at large is a large undertaking for a city, and making sure that initial goal of releasing that data set is not only met but verified is often a challenge. Releasing sets of data into the ether is only useful if it solves a current or potential problem, provides better transparency, or delivers a current service in a more affordable, useful manner.
Led by Peter Ambs, Albuquerque’s CIO, and Mark Leech, Application Development Manager, the City of Albuquerque’s open data initiative launched in 2012 with 13 data sets available to the public. Our team at APPCityLife worked with the city’s transit IT team to launch the first open data app, ABQ RIDE, which, in its initial phase, was created to provide a mobile answer to a question that was asked to 311 phone operators almost one million times a year: where is the bus? Through analytics and compiling data from a variety of sources, the city was able to determine that they had realized a savings of almost $200,000 between June and December of 2012. Since its launch in 2012, the ABQ RIDE mobile app has gained over 17,000 users on iOS and Android, with almost 1000 new users added each month.
User Interface Design: Proof of Concept
When working through the user experience design for the ABQ RIDE mobile app, one of the issues discussed in length with the city’s transit IT team was the option of constant refresh versus manual refresh of the data. The mobile website created by the University of New Mexico provided automated refresh of the data, resulting in a fun visual of the bus icons “moving” along routes as continual refreshed data calls to the transit servers regenerated the images every few seconds. It was the consensus of the design team that most of the app users would be on limited data plans through their mobile phone carriers, diminishing the usefulness of the app if it became a data-hog by forcing the app to constantly refresh when the real time tracking feature was opened. Thus, the app was released with a refresh button that would allow users to control their own data use and decide how often they wanted to refresh the data.
But the real question was whether users would intuitively understand how to access the refresh button or whether the barrier to obtaining new data would be too high. By compiling anonymous, opt-in real time analytics for specific features within the ABQ RIDE app, we were able to determine the answer to our question.
The simplest answer was, yes, users were, in fact, accessing the refresh button on a regular basis. The option to control data was working. But as is usually the case with a simple answer, the answer generated more questions. Our team was curious to see if additional analytics could answer more sophisticated questions about user behavior surrounding the implementation of the refresh option on the live tracking feature. We wanted to know where were the users located that were accessing the app, were the users within Albuquerque clustered along specific routes, and how many of those users were accessing the app within the same 24-hour window as the date of the highest number of refresh hits. We were able to determine the following:
- Over a 31-day period, the upward and downward trends between number of users accessing the app and number of times the refresh button was accessed followed similar patterns.
- On the specific date of March 12, 2014, a Wednesday, 359 users launched the ABQ RIDE mobile app. While a majority of those users were located in the Albuquerque metro area, the app was also launched in Belen, Santa Fe, and south of Gallup, New Mexico, as well as Chicago, New York and isolated locations within China.
- Users who accessed the app within the Albuquerque metro area were spread over all parts of the city with higher clusters of users located along Central Avenue and Coors Blvd.
- On March 12, 359 users refreshed live tracking data over 2,000 times.
By combining multiple sets of opt-in data, we were able to determine that the refresh option was not a barrier to users acquiring the latest bus locations – and that some users accessed the feature multiple times.
When Albuquerque’s transit IT team set out to solve a problem bus riders were experiencing, their initial goal was to create a better experience for the citizens who used transit services. And they did – bus ridership is up, and we’re told that transit calls to 311 have diminished so much that the citywide call center has reduced their operators by three positions. But what city leaders didn’t initially expect was the significant savings in tax dollars through the carefully planned launch of this single open data set. The real time mobile analytics APPCityLife was able to collect helped not only justify the initial expense for the mobile app’s development and support, but analytics continues to enable the team to enhance the mobile app user experience. Several new features for the app are currently in design, including in-app integration to 311 reporting, an interactive trip planner, and in-app mobile coupons which will provide a revenue share back to the city to further defray costs.
The marriage of open data and analytics is an incredibly powerful tool, not only for our clients, but also for our own development team. Understanding the data we collect helps our team determine what features to support on our mobile publishing platform and how to solve bigger issues on a national and international level based on the information gleaned within real-world results.