A previous article in our Data by Design series started a conversation about the importance of nonprofits maintaining data integrity, which requires data to be complete, valid and accurate. In this article, we’ll expand on this topic, discussing in more detail the importance of reliably and consistently producing high-quality data, the challenges facing nonprofits committed to that goal and the role Social Solutions plays to help them.
Nonprofits typically need to track a wide variety of data. Most funders, both governmental and philanthropic entities, are very interested in process and compliance. As a result, they establish numerous rules regarding the collection of data to make sure the money awarded is spent and reported according to their specifications.
Yet, while recognizing the significant role that data plays in providing visibility into program effectiveness to funders to ensure continued funding, many nonprofits struggle to produce high-quality data about their activities, outputs, outcomes and impact. Insufficiently staffed teams, increased pressure from funders to set up complex measurement systems, inadequate user interfaces, lack of training: those are some of the factors that create serious barriers to effective data collection in nonprofit organizations.
To further complicate matters, staff members often must choose between actually doing something about a problem and performing time-consuming data entry. Understandably, they will much prefer to invest their limited time, expertise and caring in causes and efforts in which they believe, rather than making sure all their forms have accurate and complete data.
An Investment in Good Data is an Investment in Success
The reality is that good data, used for learning, is a critical ingredient for achieving program goals and understanding successes and areas of improvement. And in times like this, in which demand for social services is soaring at the same time funding dwindles as a result of the COVID-19 epidemic, those who understand the inherent value of high-quality data will have the advantage.
Take, for example, the homeless sector. The United States Department of Housing and Urban Development (HUD) awards Homeless Assistance Grants to communities that administer housing and services at the local level and receives regular reports from the HMIS (Homeless Management Information System) applications where local information is captured. Social Solutions offers HMIS software that helps communities meet the regulatory requirements necessary for securing the critical funding they depend on. Nonprofits and agencies use our software to collect data on the provision of housing and services to homeless individuals and families and persons at risk of homelessness.
It’s not hard to see why high-quality data is a critical element in ending homelessness. Having a timely, accurate, and complete understanding of homelessness locally and nationwide is necessary for communities to provide comprehensive care to individuals and families in need, measure impact and receive funding. Without complete and correct data captured in a timely manner, housing-focused organizations can’t answer key performance questions such as:
- How long are people staying in shelters before moving to permanent housing? The shorter this length of time, the better.
- What percentage of people in shelters are moving onto permanent housing within 30, 60, 180 days? The higher the rates, the better.
- What percentage of people who left a shelter to permanent housing reenter the system within a year? The more robust the services in follow-up support, the less likely this is to occur.
Missing, incomplete and inaccurate data gets in the way of developing a full picture of performance. Without good data, it’s impossible to understand the successes achieved and the limitations the organization or program may be facing. For instance, is the length of stay in emergency shelters prior to moving to a permanent solution truly smaller for community A than for community B, or is the difference a result of more diligent data entry by community A, which has a reliable process to capture exit dates if a family “self-resolves” by finding a rental on their own?
Empowering and Amplifying Data Quality
Here at Social Solutions, we know the important role our software plays in the quality of the data produced by the organizations we serve. Our product team is regularly talking to clients to identify new opportunities to enable users to record information seamlessly and securely, minimizing effort and the risk of mistakes. And our Data Science team plays a key role in this process, partnering with many of our clients to understand their data issues and develop tools to help fix them.
To go back to our example in the homeless space, we currently have a data quality initiative to make it easier for the nonprofits using our software to find and correct data problems by highlighting missing or conflicting information in their database. As part of this initiative, we sought approval from one of our HMIS clients to look at their data to create new data integrity rules.
Examples of problems surfaced in this initiative include:
- Same family with records of move-in to permanent housing of two different types, rental with subsidy and rental without subsidy, in the same month.
- A mother with two small children identified by the same household ID but logged as separate individuals seeking housing, each listed as “head of household” in their program enrollments, when the children should have been listed as dependents.
- Individuals and families who are no longer being assisted (because they moved to another city, found housing on their own or stopped having contact for an extended period) but haven’t been exited from the program as prescribed.
As we all know, data quality issues like the ones above affect not only the homelessness sector but everyone in the nonprofit world. And it’s easy to see how they can negatively impact the ability of the organizations working to improve lives and conditions in their communities to report on the outcomes they achieve.
Incorrect or missing information can severely limit a nonprofit’s ability to learn and communicate. With good data, your organization can measure:
- Outcomes and community impact of the services provided
- Indicators of program effectiveness to present to funders to ensure continued funding
- What is working and what is not so that resources may be reallocated accordingly
- Actionable insights (in real-time) to support frontline staff decision-making
Good Data Demonstrates Impact
Helping nonprofits produce high-quality data is a key motivator that gets the Social Solutions Data Science team out of bed in the morning. As outcome expert Robert Penna points out in his book, The Nonprofit Outcomes Toolbox:
“Incorrectly entering a client’s date of birth or Social Security number at intake, for example, may, in itself, seem like a small error. But three months later, when that client repeatedly cannot access services because his identification does not match the system’s computerized records (all of which, needless to say, now contain the incorrect DOB or SS#), that initial boo-boo can be seen to have ballooned.”
High-quality data translates into more funder confidence, less guesswork, and risk in decision making, and more productive staff. And because maintaining excellent data quality is an ongoing process that requires constant vigilance, our commitment to improving the quality of the data produced by the nonprofits we serve has only gotten stronger. Despite the challenges arising during the COVID-19 crisis, we continue to see this not only as a priority but an opportunity to lead the way toward better tools to communicate the value of social services during these difficult times.
About the Author
Adriana Beal is a native of Brazil and has lived in the U.S. since 2004. Her background is in Electrical Engineering, Strategic Management of Information Systems, and Data Analytics. Prior to joining Social Solutions, she worked as a product manager for analytics products in healthcare, social media management, and cloud management, and as a data scientist in charge of machine learning models.