If you’re reading this and you work in local government, chances are you already know the importance of data. You might also feel pressure to do more with data: generate more, analyze more, report more. If this neverending push for more makes you feel overwhelmed, you’re not alone.

Every single government in the world, from small cities and counties to federal giants, is somewhere on a data journey. The farther they go, the more they’re set up to deliver effective, equitable public services. Moving forward doesn’t always happen in leaps and bounds, but even a small step in the right direction can have a big impact for residents — and there’s never been a better time. The US Treasury recommends investing State and Local Fiscal Recovery Funds to strengthen capacity to collect and analyze data and evidence, as a means of turning a one-time investment into lasting results.

Rather than try to overhaul their entire data infrastructure in a single pass, local governments can look to their peers for smaller, more attainable ways to improve their data practices. Cities in the What Works Cities (WWC) Economic Mobility Initiative, led by Results for America, set out to improve economic outcomes for residents through data-driven interventions. Along the way, many of them took another step forward on their data journey. Here’s what they did, and what other local governments can do:

• Standardize: Detroit synched address data across departments to better protect residents living in affordable housing
• Humanize: Tulsa used both qualitative and quantitative data to tailor workforce development programming specifically for disconnected youth (18–24 year olds who aren’t working and aren’t in school)
• Randomize: Rochester tested different levels of financial incentives to see how they could help low-income residents meet their savings goals
• Stay curious: All nine cities maintained a true learning mindset throughout the work, digging deeper to understand the “why?” behind the data they gathered

Standardize: How a better database can lead to better renter protection for Detroiter

In Detroit, 35% of residents live below the federal poverty line, and 1 out of every 5 households faces eviction annually. Like many cities, Detroit has policies in place to protect residents; these policies are designed not only to keep people from losing their homes, but also to keep people safe while they’re in them.

Prior to their work with the WWC Economic Mobility Initiative, Detroit housed its address data across multiple departments: Public Works, the Office of the Assessor, and the Buildings, Safety, Engineering, and Environmental Department (BSEED). However, address data was neither standardized nor shared among departments.

Detroit worked with the Center for Government Excellence (GovEx) at Johns Hopkins University to establish data governance by aligning people, processes, and platforms across the three departments. Now that they’ve taken this step, the City is better poised to:

• Reach out to landlords, which allows for more registered rental properties and more potential to integrate services for residents (for example, on-site financial counseling and/or tax preparation services in affordable housing units)
• Reach out to residents, including mailers with information about city services and prompts to report noncompliant or unsafe housing
• Enforce code compliance, including a recent push for lead safety among rental units

“There’s nothing particularly flashy about updating a database, but the downstream impact for residents is huge,” says Keegan Mahoney, Program Director at the Housing and Revitalization Department for the City of Detroit. “When cities know where rental properties are, who owns them,, and when they were last inspected, they can help keep residents safer and more secure in their homes, which can lead to better health and economic outcomes down the line.”

Learn more about how addresses have been standardized for the City of Detroit. Explore the data structure using this interactive map.

Humanize: How Tulsa connected different kinds of data to better serve disconnected youth

In 2019, there were nearly 11,000 disconnected youth in Tulsa — 18–24 year olds who aren’t in school, and aren’t working. Tulsa is not alone — some estimates put the number of disconnected youth in America at nearly five million.

Tulsa Community WorkforceAdvance (TCW) is an evidence-based, nationally recognized workforce program; NextUp is a spinoff of TCW designed specifically for 18–24 year olds. NextUp builds on the TCW model with life skills workshops, GED attainment, career exploration, and wraparound services in addition to technical training and job placement.

As part of the WWC Economic Mobility Initiative, NextUp used mixed-methods research to better understand the aspirations, barriers, and needs of disconnected youth, and design thinking to test and refine program improvements. Mixed methods research meant a blend of quantitative data (expanding their database to include more participant information so they could better identify patterns and trends) and qualitative data (structured 1:1 interviews with coaches and participants).

“We knew we couldn’t just track numbers,” says Karen Pennington, Executive Director of Madison Strategies Group, which runs TCW. “We also needed personal feedback from the people who experience our programming everyday.” Ideally, governments use quantitative data to reveal what they miss in anecdotal perceptions and biases, and qualitative data to reveal what they miss in the numbers alone.

Working with the Behavioral Insights Team (BIT), NextUp embarked on a process that was at once deeply methodical and deeply empathetic:

• Understand: Use macro-level insights from its student database, coupled with structured interviews with coaches and students, to get a holistic view of the program
• Prototype: Design resources to improve recruitment, coaching, and more, bringing coaches and participants back for feedback on early designs
• Implement: Revise and roll out
• Evaluate: Repeat a mixed-methods approach (interviews and database insights) to understand outcomes and impact

NextUp improved referrals and retention through its process, reporting a 94% satisfaction rate among graduates and a nearly three-fold increase in participation, program completion, and job placement. They also demonstrated how design thinking can lead to better public services.

“This initiative gives Tulsa the ability to further our goals of ensuring upward mobility and opportunity for all of our residents,” said Tulsa Mayor G.T. Bynum, whose dedication to growing data practices citywide has led to Tulsa’s WWC Certification. “Being able to have partners like What Works Cities has helped us use data-driven approaches to produce real-world solutions and I’m thankful for our continued partnership to be able to help move Tulsa forward.”

The initiative has also built momentum for TCW and its partners, which have received ARP dollars from the City and County of Tulsa.

Randomize: How Rochester tested strategies to support savings for low-income families

For many low-income families, tax refunds in the form of the earned income tax credit (EITC), Child Tax Credit (CTC), and stimulus payments are the household’s most significant cash infusions each year. Evidence from $aveNYC and SaveUSA suggests that saving a portion of those cash infusions can help “smooth” financial difficulties throughout the year.

“If a car breaks down unexpectedly, people have to choose between their only reliable source of transportation and paying rent — effectively a choice between losing their job or losing their home,” says Yversha Roman, who works for Creating Assets, Savings, and Hope (CASH), a local nonprofit that provides tax assistance to low-income residents. “Good savings habits can help families manage unexpected events.”

The team at CASH and the City of Rochester knew, coming into the WWC Economic Mobility Initiative, that financial incentives can help people build better savings habits. But how much is enough? It’s important for cities to know what the threshold is for incentivizing savings for residents, because they have many residents to serve, and limited operating budgets. Determining the best dollar or percentage amount to offer as a savings incentive can help even small cities participate in this kind of work.

Rochester set up a randomized, controlled trial (RCT) wherein residents saved their EITC, then received quarterly disbursements over the course of the year. These disbursements were matched by the program at 25%, 50%, or not at all (control group). While COVID-19 interrupted the work, early data suggests that a 50% match was effective, but that a 25% match didn’t change much compared to not matching at all. Participants in the 50% group saved 10% more than in the 25% group, and 54% of respondents cited the match as a reason for joining the program.

Though Rochester went on to simplify programming, their randomized trial helped shed light on what percentage match would be valuable to residents. “RCTs aren’t only for universities and think-tanks,” says Kate May, Chief Performance Officer for the City of Rochester. “For cities, counties, and tribal governments, they can be as simple as A/B testing two different ideas, interventions, or messages. Even at a small scale, randomized work can help cities quickly test hypotheses and learn what residents need.”

Stay curious: How cities made the most of their data with a learning mindset

Data-driven government isn’t just about collecting data. It’s about diving into the data with curiosity, empathy, and an open mind. Governments who use data well are always asking ‘why?’ When things do work, they’re able to pinpoint why and scale; when things don’t work they’re able to learn, pivot, and improve.

A vital way cities can stay curious is to disaggregate the data they have by demographic factors, such as race, ethnicity, age, gender, and income. This data-within-the-data helps shed light on the impact of systemic inequity, and gives cities the clarity they need to work towards better services and outcomes for all residents.

Take the next step

Evidence-based government, like evidence-based medicine, is by definition an ongoing effort. There is no future state in which governments will have exhausted or perfected their understanding of what works; therefore, no government will ever be “perfect” at data.

Rather than trying to boil the ocean, local governments should take whatever steps they’re able to: Standardize, humanize, randomize. Above all, we encourage governments to stay curious, ask ‘why,’ disaggregate data, and use ARP funds to take the next step on their data journey — because even small steps can make a big difference for residents.