Philadelphia’s Department of Revenue team starts all of its meetings by repeating one stat: 96.1% of real estate taxes owed to the city are paid the year it’s billed.

It gives the team a positive starting note, said Rebecca Lopez Kriss, a senior policy analyst and the director of strategic outreach for the department. Indeed, more of those meetings have focused on using data and tech to improve the department’s community engagement.

Some members of the team discussed the projects they’ve piloted and their payoffs at the Philly Tech Week 2019 presented by Comcast event “Data Driven — Success Stories from Philadelphia City Government” on Monday. The four panelists:

• Nathaniel Olin, social scientist at Mayor’s Policy Office
• Darryl Watson, research and information analyst
• Mark Harvey, director of operations in the Water Revenue Bureau
• Roman Strakovsky, director of data intelligence in the Department of Revenue

Here are a few ways they said the City of Philadelphia is using tech to improve its processes to better reach residents.

1. The city still relies on snail mail as a primary form of community engagement.

In fact, the Department of Revenue sent 12.9 million pieces of mail in the fiscal year of 2017.

But snail mail is not immune to improvement. Olin said the response rate to mail depended on who the sender was: Were residents more likely to attend a pop-up event if City Council sent a letter explaining it or the Department of Revenue? Perhaps unsurprisingly, City Council.

Olin worked with the department on questions like this via the PHL Participatory Design Lab, an initiative funded by the Knight Foundation’s Knight Cities Challenge grant program. It also partnered with the Office of Homeless Services to provide data-driven insights on how to improve services.o

His work with the department focused on the Owner-Occupied Payment Agreement program, which sets up affordable monthly payments for homeowners with overdue property taxes. His findings gave the department empirical evidence on which to base future community outreach, Lopez Kriss said.

“The Department of Revenue has done experiments before,” Olin said. “This is not the first time that they’ve worked on this, and this is why they were excited to do this in the first place. But just building these kinds of experiments into everything and using … reproducible workflows to build this into the city’s habits in one of the big goals.”

2. The most intimidating enforcement methods are the most effective.

It’s not surprising, but the Department of Revenue has found that the best way to hear back from residents who owe money is to take an assertive step forward like shutting their water off.

Harvey discussed the predictive analytics the department is using to prevent those situations. The revenue’s Data Warehouse, a $7.3 million upgrade to the department headed by Strakovsky, provided the Water Revenue Bureau with a desktop app that lists the delinquent accounts from most likely to pay money owed to least. Then, city workers can target those individuals and intervene before having to take further action.

3. Excel cost the city a lot of money.

Well, not exactly. But the department has seen more dues being paid once it switched from its manual data entry methods. For example, the Water Revenue Bureau would list people who owed money in an Excel spreadsheet and call them, but the data was static. It needed to be updated again and again.

The Data Warehouse app for the Water Revenue Bureau updates the list of people who owe money daily. That, plus predictive analytics, has supplied workers on call campaigns with better tools, leading to better results.

In fiscal year 2017, $2.1 million of overdue water debt was paid. It jumped to $8 million the next fiscal year, Harvey said.

Poor quality data costing money isn’t a problem specific to the city. It costs the U.S. $3.1 trillion annually, according to IBM. Watson said it’s because data is as idiosyncratic as the world it tries to capture, and people are the ones handling the data. And we make mistakes.

“Every person that comes in contact with the data during the data life process who makes a small and tiny adjustments for the things that we do to compensate for the fact that the data might not be complete or may not be accurate,” he said, “and the thing is that when we quantify all those tiny small adjustments … that’s a large dollar sign, really, in lost productivity.”

4. The Department of Revenue is, well, just one department.

Lopez Kriss said her department has been at the forefront of using data to improve its outreach and services, but it’s a general ethos of the city. (Learn more about phila.gov’s redesign here.)

The Department of Revenue is still committed to improving residents’ experiences with their local bureaucracy through accessible design. Projects to come by the department include a property owner portal, where people can log in and see how many taxes they owe, and updating forms so they’re easier to approach.

“What [people] don’t realize is that we’ve been really changing the story about collection and about delinquency,” Lopez Kriss said. “Since 2013, we’ve reduced delinquency by 31%. We have also expanded the safety net of programs for folks who might experience difficulty in the real estate chapter.”