Building a Foundation for Water Utility Data Sharing
How can we make it easier for critical data to flow?
This post was originally published as part of the Anatomy of Environmental Decision Making blog series by the Open Environmental Data Project.
From its founding, one of the core goals of the California Data Collaborative (CaDC) has been to promote the effective sharing of data among water agencies. This blog will attempt to describe the current state of data sharing for water utilities, as well as a vision and strategy to make this process easier and more transparent.
The Status Quo for Data Sharing
Right now, if you work at a local water supply agency and are asked to share data, it is likely for one of the following three situations:
1. One-time Studies and Projects
Agencies frequently share data with consultants (and to a lesser extent with researchers) to conduct studies. Despite this frequency, the process is often a pain because each consultant needs a contract in place with a non-disclosure clause, and each data request is often handled as if it were unique. In reality most people looking to access a particular type of data (e.g., billed water use) need roughly the same set of water use and customer characteristic data.
2. Reporting up the Chain
A more regular form of data sharing takes place when water supply agencies are asked or required to share a particular set of data fields with another government agency. This may be their wholesale water supplier requesting data on an annual basis so that the wholesaler can conduct their annual planning process. More frequently, it is one of the state water agencies requiring a suite of regulatory reports at monthly, annual, or quinquennial intervals.
These reports require local agency staff to act as a translation layer, taking the raw information their agency collects and grouping, filtering, and aggregating it into the format expected by the recipient. Apart from being a tremendous time-sink, this process is also slow and prone to error.
3. Sharing with Software Vendors
Possibly the most successful data sharing model yet is the way that water agencies are generally able to deliver timely and consistent data to the software vendors that provide them with an essential service. A few examples we are familiar with include online portals that allow customers to view their water use, vendors that generate and format paper bills, and software for data analysis that support key workflows (e.g., leak alerts).
Some of this success is due to the configurability of software and the power of automation, but perhaps the most important factor is that the agencies rely on software vendors to provide them and their customers with mission-critical services, so they are highly incentivized to get data sharing right. When customer bills and communications are on the line, the data had better flow!
A New Path for Water Utility Data Sharing
This brings us to the strategy that the CaDC is currently pursuing to improve water utility data sharing.
1. Build Fundamentally Useful Tools on Top of the Data
Rather than building a data sharing product and assuming that agencies will put their data in it, it makes more sense to adopt the successful model of software vendors and build a tool that agencies really want to populate with high quality and accurate data. To do this, you need to make it as easy as possible for users to get their data into the tool, and you need to ensure that the tool is delivering real, and ideally essential, value to the user.
2. Make it Really Easy to Securely Share the Right Data
Many of the software systems that are great at delivering essential services to water agencies are currently not great at enabling easy access to that data for sharing. At best, they have little incentive to make data sharing easy;at worst, they may have an explicit incentive to make data export difficult to lock users into the vendor’s proprietary ecosystem. This is often the case with the systems of record that many agencies rely on.
We need a new model where the APIs, easy exports, and off-the-shelf integrations common in consumer software make it into core water utility systems.
3. Operate as a Trusted Steward of the Data
For a data sharing strategy to succeed, both the organization and the platform need to have the user’s trust that their data isn’t going to be sold to the highest bidder or sent off to a regulator without their consent. This can be accomplished through a novel business model (non-profit vs. for-profit), strong governance, and transparent communications.
With these pieces in place, one can begin to see the outlines of a trusted, secure, and sustainable model for streamlining the movement of mission-critical water data around the industry to aid in environmental decision making.