This guide is designed to help cities understand what a data inventory is and how to inventory data. It begins by explaining the basic concepts and value of a data inventory. Then, it features insights from city employees who have gone through the inventory process. Finally, it provides links to city resources, guides, and templates of actual data inventories.
What Is a Data Inventory?
A data inventory is a fully described record of the data assets maintained by a city. The inventory records basic information about a data asset including its name, contents, update frequency, use license, owner/maintainer, privacy considerations, data source, and other relevant details. The details about a dataset are known as metadata.
Because cities may have thousands of datasets across multiple servers, databases, and computers, it’s helpful to narrow down which datasets should be included in the inventory overall and how to plan for inventory updates in the future. The datasets worth inventorying are those which are considered assets to employees, departments, executive leadership, and the general public. Data assets can range from individual datasets that are connected to forms that people fill out, to integrated databases that track a city’s operations in any given field (building permits, public safety responses, etc.)
Why Conduct an Inventory?
The first step to treating your city’s data as an asset is to create a comprehensive data inventory with consistent metadata. Knowing what data your city collects leads to efficiency, and increases accountability. It also eases citywide reporting, decision making, and performance optimization.
Managing a data inventory reduces risk and uncertainty by creating a checklist for security and compliance requirements and improves a city’s ability to designate accountability for the quality of the data collected and created. Just as it is important for cities to know what data they have, it’s equally important to know what data a city does not have. With a complete picture, cities can begin to collect and use city data to better align mission goals, increase consistency and confidence in decision making, and build performance intelligence.
Managing a data inventory is crucial to better information sharing and integration and a sustainable comprehensive open data program. Providing a public data inventory will make city employees’ jobs easier when they need information from another department - they will know what exists and how to find it. The same benefits apply to the public regarding its search for city information. Having a complete inventory is also important when determining which datasets to release publicly. It’s not feasible to release all of a city’s public datasets at once, so decisionmakers need a prioritization strategy. The data inventory can be used to prioritize the release of data according to strategic priorities, public interest, etc.
How to Conduct an Inventory
The data inventory process is carried out in the following steps:
Step 1: Establish an Oversight Authority - Conducting a data inventory across departments requires coordination, oversight, and leadership. The first step to conducting an inventory is establishing who will manage the inventory process. Oversight authorities can come in a variety of shapes and sizes and are often defined in a city’s Open Data Policy. Some are led by a Chief Data Officer (or similar role), others leverage existing enterprise data management bodies, and others are working groups that include public representation. While the breadth and depth of data governance authorities can range to best suit your city’s needs, establishing a clear authority body to oversee the data inventory process is key to success. In the absence of a data governance committee, you may wish to identify a lead liaison, preferably within the Mayor or City Manager’s office, to interact with departments and facilitate this process.
Note: Establishing a data governance committee, or repurposing an existing committee, is an optional, but highly recommended step in successfully completing the inventory process.
Step 2: Determine the Data Inventory Scope and Plan - The oversight authority, such as a data governance committee, should manage the inventory process by providing an unambiguous scope, deadlines, performance metrics, and guidelines.
Scope: If the scope is not already defined in your city’s Open Data Policy, the oversight authority should determine the scope of the data inventory at hand. If your city does not already have a data inventory in place, creating a city-wide comprehensive data inventory can range in difficulty depending on how many data assets your city manages, how siloed those assets are managed, and your available capacity to conduct the inventory. When defining the scope of the data inventory, the oversight authority should consider the following:
Plan: The data inventory plan.
Step 3: Catalog Data Assets in Accordance with Inventory Plan - Liaisons in each city department or agency catalogue and describe the data assets within their departments. Liaisons are employees who are responsible for managing the inventory process at the departmental/agency level. The lead manager of the data inventory compiles the individual department inventories into a larger citywide data inventory. Inventories should be structured in machine-readable format (Spreadsheet, CSV, JSON, etc.)
Step 4: Data Inventory Quality Checks
Step 5: Initiate Data Prioritization Efforts - The data governance committee establishes the priority and scheduling of the publication of datasets described in the inventory.
GovEx has helped a variety of city governments work through the inventorying process. When creating this resource, we reached out to many of those city champions and some seasoned veterans to surface best practices, learn about opportunities for improvement, and dive deeper into what worked well. The following is a list of takeaways from those conversations:
Governments complete data inventories for a variety of reasons. Data inventories are a great way to figure out what data is being collected (and if there is any duplication among departments), determine what systems are in use and their analytics capabilities, promote transparency, develop data publishing plans, and learn about current challenges and opportunities within the organization that might affect its open data goals. Because many inventorying efforts require participation from a wide variety of staff, inventorying is also a great opportunity to build relationships and convey the importance of inventorying and the open data program.
You don’t need an open data policy to complete and find value in a data inventory. GovEx surveyed municipalities with and without open data inventories, the consensus is that having an open data policy that calls for a data inventory is helpful in completing the inventory in a timely manner and demonstrating its importance across the organization, but not necessary.
There’s no one size fits all approach for data inventories. Inventories should be customized to fit the government’s needs and open data goals. Some governments begin with a targeted approach to inventorying in one department, one IT system, or around one strategic priority; other governments dive right in attempting to inventory all their data systems and datasets in one go. Some governments have their open data coordinators complete the inventory; others hire third-party auditors. It’s important to take time to determine what the right approach is for your organization. This includes exploring how familiar staff are with open data, how bought in they are to your organization’s open data goals, their capacity to assist in completing an inventory, any open data legislation that might deal with inventorying, and how you plan to share the results of the data inventory.
Inventorying works best as an organization-wide effort. Inventorying data is a chance to connect with city staff, relay the importance of the city’s open data program, and provide training around open data. Making this a citywide effort can be a unifying process that thoroughly addresses concerns about open data, builds buy-in throughout the city, and generates conversation among frontline staff, managers, and senior leadership about data.
Training is the first step to creating a good inventorying experience. Members of your organization undoubtedly have different understandings and knowledge about open data and the importance of completing an inventory. Providing information about open data in general, the city’s open data program, its goals, and why it’s doing an inventory ensures that everyone is on the same page and motivated to contribute to the inventorying process.
Inventorying is a continual process. Some cities have mandates which require them to update their inventories on an annual basis, but all the local government we spoke with plan to update their inventory routinely and regularly.
A note about privacy. Do not exclude any datasets based on privacy or confidentiality concerns. To make the data inventory as useful as possible, it should include data that may be sensitive, private, or unlikely to be released. Always include a description of the sensitivity concerns.
When it comes to inventorying, there’s no need to reinvent the wheel. A lot of cities have gone through the inventory process and publicly share their resources. The following list includes data inventory guides, templates, publishing plans, presentations, publicly released inventories, and workflow diagrams for completing inventories from cities across the United States. There are many ways to structure and scope a data inventory; the first step is deciding what works best for your city.
Step 1: Establish an Oversight Authority
Step 2: Determine the Data Inventory Scope and Plan