Noah Urban is a Senior Analyst at Data Driven Detroit (D3), a low-profit social enterprise that works to help partners use data to drive informed decision-making in Metro Detroit. At D3, Noah manages execution on D3’s suite of projects and initiatives, including a number of projects with Microsoft. He joins us today to discuss one of those projects — a data collaborative tentatively called the Metro Detroit Data Alliance, or MDDA.
At Data Driven Detroit (D3), we love data, and we love that data allows us to make more informed decisions about the world around us. That said, we also know that data can be difficult to understand, use, and access, and can sometimes tell wildly conflicting stories. We’ve written plenty about this in the past, and we’re not alone. Ask three different organizations in Detroit how much investment has occurred in the city over the past several years, and you’ll get three different numbers. This fragmentation of information makes it harder for people to coordinate their work effectively.
A data collaborative can help to address this problem. The idea behind it is fairly simple (in theory, at least – we’ll get to implementation in a moment). First, local organizations (governments, nonprofits, universities, and even private companies) sign up to contribute information into a central data repository. Next, these contributions are automated by developers staffing the collaborative so that the organizations sharing data don’t have to change their day-to-day workflows. In an ideal situation, partners barely have to commit any time to sharing with the collaborative, and updates to data will occur automatically. Finally, the data are reported out to the public through a central interface that can take a number of forms, including an API, a data portal, an online library, or some combination of all of these. The value of the concept is pretty clear – the data collaborative integrates information from multiple providers working in the same space. This increases the quality and comprehensiveness of the data, helping to ensure that everyone is working off of the same set of facts.
While the concept is pretty straightforward, implementing a data collaborative is incredibly complex. It involves enormous investments in technology and staff time, plenty of careful and thoughtful partnership building, and no shortage of legal reviews of frameworks and memorandums of understanding. Because of this, few cities have managed to even get this sort of initiative off the ground. One of the best examples of a regional data collaborative is the Western Pennsylvania Regional Data Center in Pittsburgh, but this is definitely still a developing field of practice with few examples of long-term success. The cost of building out the technical infrastructure to support a data collaborative is one challenge, but perhaps even more difficult are the tasks of convincing local partners of the value of sharing their information into this centralized system and addressing concerns or insecurities that partners may have around sharing imperfect or incomplete information.
In Detroit, we’re lucky to have a fantastic network of partners supporting Data Driven Detroit in implementing a prototype data collaborative to demonstrate the value and power of the concept in practice for Southeast Michigan. Microsoft has generously provided support to build out an Azure-based API that allows developers to connect into D3’s prototype data warehouse and query the information within. D3 has also joined with a coalition of partners, including the Downtown Detroit Partnership, the University of Michigan, the City of Detroit Department of Innovation and Technology, the Detroit Riverfront Conservancy, and the Detroit Economic Growth Corporation (among others) to develop a prototype using locally-sourced data for the Downtown Detroit area. Our hope is to use this prototype to build the case for a fully-fledged regional data collaborative in Southeast Michigan, helping to ensure that in the future, we can all make better decisions and report our impacts more consistently.
Our next steps in this journey include building out a prototype front-end to show the sort of information that a collaborative like this can provide to the world, and continuing to work with our partners to incorporate more data into the central warehouse. We hope you’ll continue to follow our data collaborative effort as it moves forward! Sending a special thanks to Microsoft, both for their considerable support of the project thus far, as well as for extending this platform for us to share the idea!