Beyond Transparency: Louisville's Strategic Use of Data to Drive Continuous Improvement

Government embodies our highest aspirations and our lowest expectations.

We have high expectations for all that we want our government to do and all of the services we ask it to provide, regardless of politics or fiscal constraints. We expect government to collect our trash, inspect our restaurants, maintain our roads, protect our property, and provide services to our most vulnerable citizens. These are just a few of the myriad tasks. Most governments perform these functions with a level of resource and reporting constraints that would challenge any well-run organization. Many of us have experienced, in one way or another, the reality that the majority of cities do not operate like well-run organizations.

Much has been written about the principles and practices of creating great organizations. In Louisville, we’ve focused on how to implement those principles and practices in the government context, using open data to drive a culture of continuous improvement.

Government Is Not Special

Whether it’s getting a driver’s license, ordering a new recycling or trash bin, or reporting a pothole in the road, all too often, citizens experience an overwhelming inefficiency in the process. The whole encounter is too long, overly cumbersome, or unnecessarily redundant. The customer (citizens, businesses, academics, etc.) frequently leaves, frustrated and potentially angry about his or her experience. These experiences are too often excused by everyone, customer and government employee alike, as “that’s just government for you.”

As a society, we are stuck in the mindset that government is somehow different and unique from private or social sector counterparts. We assume that any of the practices they employ to address complex challenges couldn’t work or wouldn’t apply at a city government level. This is simply not the case.

We’ve been inspired by contemporary books like James C. Collins’ Good to Great **(2001), Ken Miller’s We Don’t Make Widgets **(2006), and more recently, Eric Ries’ The Lean Startup** (2011). They’ve taught us that government can be more like the best parts of a successful business or new startup—innovative, proactive, fast moving, and responsive to the needs of its customers. Government can perform on par with the best-run organizations in the world, and in Louisville, we are working hard to prove it.

Over the last year, Louisville Metro Government has:

• Removed more than two hundred days from key administrative processes, like hiring.
• Reduced unscheduled overtime and workers’ compensation expenditures by more than $2 million.
• Certified more than one hundred employees in proven performance improvement methods for measuring, tracking, and improving results in an organization that are widely used in the private sector, such as Lean, Six Sigma, and Project Management.
• Better aligned the budgeting process with city and departmental strategic objectives.
• Continually evaluated the performance of multiple, unique departments and shared the results of those evaluations with citizens in a comprehensible, online format.

And, the Louisville Metro Government’s Office of Performance Improvement (OPI) helped make all this happen with a full-time staff of just three people and a budget of no more than $300,000. These accomplishments have been possible because of the use of open data and some basic startup and business innovation principles.

For cities that have begun to use data and best-in-class management and performance improvement practices, these results are not unusual. The City of Baltimore implemented CitiStat, a comprehensive data tracking and performance management tool, in 2000. The program was designed to maximize accountability by requiring city agencies to provide CitiStat analysts with metrics representing performance. During monthly and bimonthly meetings with the Office of the Mayor, each agency must examine sub-standard performance and propose solutions that can be carried out in an efficient manner. By 2007, they had used data to strategically reduce and control worker’s overtime to save more than $30.9 million. The program has been so effective that it has been sustained through three different administrations and supported by both sides of the political aisle. After leaving city office, former Baltimore Mayor (now Maryland Governor) O’Malley applied the CitiStat principles to create Maryland’s StateStat performance improvement program with equal success.

In Iowa, the state government has used Lean process improvement methods in the last three administrations. It’s not about politics, but about “doing government better,” says Teresa Hay-McMahon, president of the Iowa Lean Consortium, a group of Iowa businesses and governments that employ Lean to drive improvements in time, cost, and quality.

While many cities succeed with management practices (like Lean, Six Sigma, and Performance Management) and measurement systems (like CitiStat or StateStat), many other cities have attempted these improvements and given up. So, the questions become: How do you successfully apply these concepts to the work of city government? What does it take to make change stick? How do you get the buy-in you need to sustain it and really make an impact for citizens and your bottom line? How do you get beyond data for the sake of data and use it to inform decisions and drive culture change? We’re tackling these questions in Louisville and learning some lessons along the way.

Open Data is the Foundation

In January 2003, the City of Louisville and the surrounding Jefferson County merged to form a consolidated city-county government. Today, the Mayor and a twenty-six-member city legislature called the Metro Council govern Louisville Metro.

While the city population is approximately 750,000 people, the combined city and county population includes close to 1.3 million people, making Louisville the thirteenth largest city in the country. Nearly 5,500 dedicated employees work across more than twenty departments and agencies to service the city and county. The workforce is largely unionized, with approximately seventy-six percent of all employees in a union. The city’s general fund budget is approximately $530 million, and the total operating budget is just over $650 million.

The Start of Our Journey: Financial Transparency

Louisville’s journey from transparency to truly open data started in 2009, when the Metro Council passed a financial transparency ordinance. The ordinance established the guidelines that the transparency site was to meet. It established that the Metro Council would begin to publish detailed expenditure information for the various programs in the Metro Government.

The first iteration of the website featured information including employee names and salaries, annual budget and funding sources, quarterly revenue estimates, and annual audits. We also published all metro payments for supplies, personnel, equipment, etc., including a description of expenditures with links to actual grants or contracts, and all contracts valued at $50,000 or more.

One of the discussions surrounding transparency during the previous administration was how to help citizens interpret the data. There was some concern that just putting information on the transparency site without context would lead to misinterpretation.

What we feared most happened. The transparency site went live, and the next day, there was an article in the blogosphere about how Metro Government had spent $28,000 on alcohol. Those who worked on the site were scrambling to explain the expenditure. The reason was simple: Metro Government licenses liquor establishments, and the amount showing in the checkbook was the license refunds for the year. But releasing that data without context or explaining the license refunds created a needless media headache.

Making Open Data Serve the Citizen

We learned two lessons from that experience: the data needs user-friendly names, and it needs to be released in context to avoid misinterpretation. Louisville’s Open Data Portal has come a long way from that initial launch, and we’ve made efforts to present the data in a way that puts citizens’ needs first. As of 2013, the new and improved site provides anytime access and views into the latest city data in an easy-to-use format. It now features the most popular datasets for download in an easy-to-interpret interface, with more options to view, sort, and filter—great for anyone wanting to see city data and for IT developers wanting to build apps or online tools in creative new ways using city data.

We offer raw datasets for city expenditures, employee salaries, board members, park locations and amenities, animal services population results, news and calendar events, and links to other sites with useful data and maps. Accurate data is automatically uploaded, eliminating resource- and time-intensive manual processes. Most importantly, published data sources are exposed via API with metadata and filtering views, giving software developers more flexibility in how they can use the data.

Even with all those improvements, however, there was still more to be done to leverage data better internally in order to provide transparency into city operations, give citizens access to use that data to build new tools, and create a more efficient city that delivers better services and more effectively allocates taxpayer dollars.

Moving Beyond Open Data to Continuous Improvement

Louisville’s early open data efforts had already been underway for several years when Mayor Greg Fischer took office in January 2011. The mayor put forth a Citizen’s Bill of Rights that underscored the importance of the city’s ongoing focus on transparency and open data. Transparency was one of six main priorities. The mayor recognized that it could be the key to delivering on other commitments.

Mayor Fischer is the driving force behind the continuous improvement movement now taking root across Metro Government. He is a former entrepreneur and successful businessman who helped launch a Center for Quality of Management (CQM) chapter in Louisville for business owners in the 1990s. He came into office on the promise to run the city government more like a successful business. He started by asking questions that are elementary in successful businesses: What are we trying to accomplish? What do our customers (citizens) expect and need from us? Where are we doing well and where is there room for improvement?

Mayor Fischer was familiar with Baltimore’s CitiStat program and wanted to launch a similar program in Louisville. To run this program, he created a new role titled Chief of Performance Improvement. The role is one that many for-profit companies have also created, though it’s more often called CPO or Chief Performance Officer.

The Chief of Performance Improvement began with the mandate of helping departments answer the question “What are you trying to accomplish and how would you know if you’ve done so successfully?” Once that question was answered, the city could focus on being sure that each department had the skills and resources to accomplish its mission. This led to the three key areas of focus: planning (what is the city government doing today and what does it want to do tomorrow?), performance management (how well are we doing it?), and continuous improvement (how do we do the work and how can we do it better?).

Alongside the new Chief of Performance Improvement position, the mayor launched Louisville’s Office of Performance Improvement in January 2012. That same month, with the support of the Technology Department, the first Louisville Statistics (LouieStat) forum was held with the Department of Public Works & Assets.

The premise of LouieStat is that each participating department identifies and tracks a series of success metrics specific to its work against its own history, goals, and relevant benchmarks. Additionally, the entire city is tracking certain enterprise metrics: unscheduled overtime, workers’ compensation claims, complaints into MetroCall 311, and the amount of sick leave used by employees. Every six to eight weeks, the department meets with the city’s entire senior leadership team, including the mayor, to go through these numbers, identify issues, and address key problems.

As Mayor Fischer admitted, getting started wasn’t easy:

The first forum with Public Works and Assets was pretty rough…The department came with twenty-six pages of data and metrics, and the senior leadership team had no way of interpreting the data as good, bad, or indifferent. (G. Fischer, personal communication, 2013)

The forum, however, proved to be a critical first step in improving the way the government works. First, it sent the message that we weren’t going to wait to develop something perfect, but that we were going to get started and improve as we went along—“Sixty percent and go,” as Mayor Fischer would say. For an entrepreneur, this is known as the MVP or “minimum viable product” concept (Ries, 2011). Once you have an MVP—the most bare bones version of what you’re trying to build—you launch it, test it, and immediately set to work improving it incrementally, rather than making big upfront investments to perfect something that may or may not work.

The forum also made it clear that the focus was not just on data. It was about converting data into useful information to drive issue identification, problem solving, decision-making, resource allocation, and, ultimately, performance improvement. Data in a vacuum is never useful, so while we had a good amount of data in raw form, we needed to start using the data to identify ways to continually improve what we were delivering to citizens and how.

Implementation Is Messy

In that first year, OPI brought nine departments into the LouieStat program: Public Works & Assets, Fire & Rescue, Corrections, Parks, Public Health & Wellness, Animal Services, Codes & Regulations, EMS, and Economic Growth & Innovation. Departments were prioritized based on criteria including size of department, citizen impact, performance readiness, and span of control.

As part of integrating LouieStat, departments began working with OPI analysts to develop the right operational metrics for their specific functional area and to put together historical data and relevant benchmarks. This was complicated, since some departments, like Fire & Rescue, had a wealth of operational data, and others, like Parks, had very little data—most of which was captured in handwritten paper logs.

As we began to track, analyze, and share the data with departments, all of the issues one might anticipate began to surface. Many departments did not have a strategic planning process, and as a result, had never clearly articulated their goals externally or internally. For those who had or were able to create goals, there was still a great deal of resistance to creating and evaluating metrics. Many complained that the data was not good; that it didn’t represent the true situation; that it was misinterpreted; or that there were inappropriate entries or people being counted “against them” in the system. All of these complaints were valid.

As we had learned in our initial open data work, though, the data will never be perfect, at least in the beginning. The minute you start tracking and analyzing it, you will begin the process of improving it—so don’t wait for the data to be perfect to start.

Still, we quickly realized we needed a way to focus in on the data that mattered most to the mission success of each department.

Strategic Planning in Performance Management

Strategic planning needed to be a part of our approach if we were going to focus our analysis on what truly mattered. At that time, Metro did not have a comprehensive or coordinated strategic planning process. While some departments had their own planning processes in place, many did not, and strategic and operational planning were lumped into and driven by the annual budget process. OPI helped establish a new planning process. The goal was to translate the mayor’s multi-year vision and objectives into a comprehensive strategic plan that cascaded throughout Metro Government. OPI also helped align the strategic goals and initiatives of all Metro departments and agencies with the current administration’s goals.

The process culminated in the mayor’s six-year strategic plan for Louisville Metro Government and more than twenty individual six-year strategic plans at the department or agency level. Contained within each plan are measurable targets we are trying to achieve as a city and as individual departments, so we can check our progress as we go and focus our attention on what matters.

We are now in the process of designing a type of “budgeting for outcomes” process that should align the city’s spending with its priorities. This design process is already raising important questions about the role of government in achieving the city’s priorities. At the very least, connecting strategy to budget is forcing departments to question the value of some entrenched programs with unclear outcomes.

Performance Management: How Are We Doing?

The strategic plans gave us a good understanding of what we were trying to do at a strategic level, and to some extent, at an operational level. Yet, we were still faced with a wealth of potential metrics for any given department. Public Works & Assets, which has five major divisions, including Solid Waste Management, Fleet, Facilities, Engineering, and Operations & Maintenance, has hundreds of operational metrics we could evaluate. The challenge is finding the vital few “Key Performance Indicators” (KPIs). OPI worked with departments to establish the best KPIs for their work, focusing on answering two critical questions:

1. What results (informed by strategic plans) are we trying to achieve?

2. How would we know if we were achieving them?

For example, we established three KPIs for the Louisville Fire Department: amount of property damage (in dollars), number of civilian fire injuries, and number of home fire inspections. We set goals for each KPI informed by past performance and where we want to be—$7,717,608 in property damage, zero civilian fire injuries, and 1,032 home fire inspections completed—and track and report progress related to those goals.

The departments in LouieStat come before the mayor and his senior leadership team (which consists of the heads of Finance, HR, Technology, Legal, Policy, Performance Improvement, and Communication) every six to eight weeks. Through consistent tracking and data analysis of their KPIs, the team works to identify and discuss what the department (and Metro Government) can do to continually improve the services it delivers to the citizens of Louisville. This also means ways we can better meet the strategic goals we’ve committed to. In between LouieStat forums, OPI supports the departments through measurement identification, data analysis, performance reporting, training, and coaching. All KPIs are then posted on the public LouieStat website (www.louiestat.louisvilleky.gov) for citizens to view.

The LouieStat process faced much resentment and skepticism at the beginning, but once departments got going, attitudes started to change. While the focus is on where the department is underperforming—and identification of the root causes impacting performance can be quite targeted—the mayor frames the interaction with each department as ultimately asking the question “how can we help?” This has been a refreshing surprise to many department directors. Because of LouieStat, department leaders gain important insights that offer them a clearer direction in managing employees and a quantifiable measure of their successes.

This experience led to a realization of how important it is to celebrate success. For many, simply being able to present the work of their departments in a meaningful and clear way to the city’s senior leadership team and the mayor was a powerful incentive.

Mark Bolton, Director of Corrections, explained:

As you go down a path of something new, there’s always trepidation. What is this about? What are we getting into now? But it truly is all about how can we be the best at what we do…that’s what Mayor Fischer is all about and that’s what this department is all about. (Bolton, personal communication, 2013)

For those departments not yet participating in LouieStat, Louisville Metro Government created a “Day of Celebration” during which teams nominated from across city government were recognized for their innovations and successes. The inaugural Day of Celebration in 2012 was hosted as a part of Louisville’s internationally recognized IdeaFestival and celebrated both performance improvement and innovations in government. More than a hundred nominations were submitted, and recognition was given to employees at all levels of the organization for contributions to “daily, continuous improvement and breakthrough work.” In addition to recognition of success, the annual Day of Celebration includes a training and education focus with breakout sessions providing an introduction to performance management tools.

Capability Building and Continuous Improvement: How Do We Go From Here to There?

Don’t just give me your hands, give me your hearts and minds.

—Louisville Mayor Greg Fischer

Many transparency initiatives and data programs stop short of their true potential, equating putting the data out there with success. OPI even started with the mantra “What gets measured, gets improved.” However, we realized that this is an oversimplification. It is not sufficient to simply evaluate current performance, set a target for where you want the department or organization to go, and trust that it will get there on its own.

As recent literature in behavioral economics makes clear, most people inherently want to be good, if not great, at what they do. However, many people, especially in municipal government, have not been supported with the training or resources to do so. The culture has not been set up to support continuous improvement. Many municipal employees are operating with legacy systems, and in some cases, legacy mindsets. To be successful, a performance management program must focus on embedding the skills and capabilities required to improve performance and truly close the gap between baseline data and the targets or goals established for each KPI within departments. Transforming to a high performing, continuous improvement-focused government is no easy task.

What it really comes down to is changing the mindsets and behaviors of those in government to recognize the opportunities around them and have the know-how and ability to then capitalize on those opportunities to deliver positive change. Of course, this is much easier said than done. Employees need to change their habits and behaviors, and the organization needs to adopt new processes.

Many people come to work every day and are completely occupied with “fighting fires”—rushing to battle crises as they come up. They work straight through, get to the end of the day, and aren’t really sure what they have to show for it. They know they’ve been busy, but they don’t feel like they’ve made any progress. Others come to work and do the same thing day in and day out. They do the work given to them or asked of them without really enjoying it or taking pride in it. We have their hands, but not their hearts and minds.

We want to help people get out of the business of fighting fires (as much as we can, given the nature of city services and municipal government) and only using their hands. We want to get them into the habit of using their hearts and minds to challenge the status quo and continually improve what they deliver to citizens and how they deliver it.

Over the course of our effort, we’ve identified three general categories of the way that people react to change. There are those who relish it, those who are skeptical of it, and those who truly dislike it:

• Those who relish change find it motivating and exciting, are always open to it, and are often the first to “buy in.”
• Those who are skeptical of change are not convinced of the value of the change being made or they have seen many unsuccessful attempts at change. They want to be sure the effort will be successful before they “buy in.”
• Those who truly dislike change only see it as extra work and are happy with the status quo. They will be the last to come around and may never truly “buy in” to the change.

Our focus at the start of our journey has been to engage those who relish change. We see them as the early adopters or “first followers,” if you are familiar with the “Dancing Guy” YouTube video on how to start a movement (if not, we highly recommend you check it out). Not surprisingly, the proportion of people in the different groups follows the normal distribution of the bell curve, with those who embrace change first equal to about twenty percent of people in the organization. They can then be used to help bring along the approximately sixty percent who are skeptical of change. The twenty percent remaining who truly dislike change will either eventually join in or perhaps leave the organization.

Those skeptical of change often have a good reason to be. Many change efforts or transformations fail. They come in with a new administration and are gone when the administration leaves or, worse, before they ever get off the ground. One of the better explanations of complex change management defines five key elements for successful change management: vision, skills, incentive, resources, and an action plan (Villa and Thousand, 1995). The absence of any one of those core elements yields something less than successful change. In Louisville, we’ve distilled the above components into three principles and added a fourth to anchor our efforts to continually improve government:

• Communication, Vision, and Action Plans: “I know what is expected of me, I agree with it, and it is meaningful.”
• Building Reinforcement Mechanisms and Incentives: “The structures, processes, and systems reinforce the change in behavior I am being asked to make.”
• Developing Skills and Offering Resources: “I have the skills, capabilities, and opportunities to perform in the new way.”
• Driving Culture Change through Role Models: “I see leaders, peers, and others behaving in line with the changes requested of me.”

With these four key principles in mind, we began engaging the group of first followers by communicating. Within a month of launching OPI, we sent an email out to all Metro employees about the work of OPI and asked if anyone was interested in being involved, sharing thoughts, and learning more about Performance Improvement. In the first couple of months, we had over one hundred responses. This was not a huge number of people for an organization of nearly six thousand, but enough to get started and help scale the impact of an office of three people. For these individuals, as well as Metro leadership and department management, OPI offered training in Lean, Six Sigma, Project Management, and overall management best practices. Many of these individuals were then assigned to cross-functional teams.

OPI facilitates the work of cross-functional teams who are asked to solve known problems that span multiple departments or stakeholders within Metro Government. Their job is to come up with plans to directly support a strategic objective or goal. In the first year, teams addressed issues impacting our structural budget imbalance, like high unscheduled overtime, long hiring processes, and inappropriate cost recoupment for special events.

This Stuff Really Works: Tangible Results

Since the launch of OPI in 2012, Louisville Metro Government and the various departments and employees involved in the continuous improvement work have reached a number of accomplishments.

Public Works & Assets: Missed Trash Pickup

Public Works & Assets was the first department in the LouieStat program. Citizen calls for service through the MetroCall 311 system was one of the initial enterprise metrics evaluated. The most prevalent calls into our 311 system (approximately six hundred calls a month) were about missed trash pickups. PW&A came to an early LouieStat forum with an “analysis” of their data and a plan to improve their performance. The department’s proposed solution was threefold: to buy another truck, hire more people, and reassess their routes. Each of these represented a significant cost. While the department was basing their recommendation on data, it brought up some vital questions:

• Did six hundred missed trash pickups equal poor performance?
• Were the missed pickups due to resource constraints, poor routes, or something else?
• If we made the recommended changes, what level of improvement could we expect?

As OPI worked with the department to answer these questions, the six hundred missed trash pickups per month shifted from being perceived as a poor result to actually a successful result. PW&A conducts over 800,000 trash pickups a month. Six hundred missed pickups is a miss (or error) rate of less than one percent! Statistically, this is quite good. When we compared the “success” of this metric with some of the other issues competing for resources in PW&A, we could not justify the investment it would take to close the gap on a small rate of error—not when we had roads to pave, sidewalks to fix, and other issues to address. Understanding the data underlying the perceived problem helped us better allocate resources where they were needed most.

Corrections: Fingerprinting Errors

At any one time, Louisville Metro Corrections department may have anywhere from 1,700 to 2,200 inmates in its facility. There are numerous processes that support the safe intake, housing, and release of these inmates. Fingerprinting is one of the most important processes. Completed at the time of an inmate’s booking, fingerprints help identify the inmates in our custody. Fingerprints are sent to the state for verification, and any incomplete or improperly taken prints are sent back to Louisville to be retaken and resubmitted.

Each month, corrections would get anywhere from two to three hundred fingerprints sent back from the state for reprocessing. This required the inmate to be pulled from housing, taken back to the booking area, and reprinted. This level of rework was inefficient and added an unnecessary level of risk. Applying a reactive problem solving process, the corrections leadership team began by identifying the problem and analyzing the root causes. As they looked into the data, they found that the fingerprints coming back were all taken by a handful of officers who had never been formally trained on how to fingerprint. When they evaluated the booking process, they found that no single officer was assigned to fingerprinting. Instead, whoever was in the area when an inmate came in took the fingerprints. With this information in hand, the department began a training program for booking officers on how to properly administer fingerprints. The department assigned two trained officers each watch to administer fingerprints. Within one month, the number of fingerprints sent back from the state dropped from an average of 250 per month to 10.

Louisville Metro Government: High Cost of Unscheduled Overtime

In January 2012, Louisville’s Chief Financial Officer and Director of Human Resources produced a report that revealed that more than one in five city employees increase their base pay each year by at least fifteen percent with overtime—and some city employees were doubling their salaries. In total, the city was spending $23 million on overtime each year. Scheduled overtime, which is either contained in collective bargaining agreements or through state laws, accounts for about thirty-two percent of overtime, or $7.2 million.

Unscheduled overtime accounted for nearly $14 million and was spread across most city departments. We used LouieStat to track the amount of unscheduled overtime a department pays out and formed a cross-functional team to work on the problem. Through the reactive problem-solving process, the team identified the largest drivers of unscheduled overtime and targeted solutions within departments, as well as across Metro, to reduce the amount. Solutions included using a redesigned tracking and approval process within departments and changing some of the language in union contracts. We also generated a new monthly report for department directors that shows just how much overtime is being paid out in their department. It identifies the supervisors approving overtime, the departments’ current totals compared to previous year totals, and estimated overtime budgets. With these changes, Metro has seen a reduction in unscheduled overtime by more than $1.4 million, or fourteen percent, in the last fiscal year.

Louisville Metro Government:

Lengthy Hiring Process Cycle Time

Complaints over the time it took to fill vacancies were high in January 2012. The hiring process could take anywhere from two to six months—if you were lucky. This length of time was attributed as one of the driving factors of unscheduled overtime costs. As the cross-functional team applied PDCA and Lean to the issue, they quickly mapped out a process that could take anywhere from twenty-eight to more than three hundred business days to fill a vacancy—over a year for some positions.

By analyzing the steps in the process for the department, Human Resources, Finance, and the Mayor’s office, Metro discovered opportunities to streamline and improve it. For example, departments were required to get OMB’s approval to refill a position if someone resigned or was fired during the year, even if the position had already been approved and budgeted for. By cutting redundant steps in the process, multiple signatures from the same office, unnecessary approvals, and idle wait time between steps, and instituting more structure to the process (i.e., standardizing the process across departments, placing time limits on how long any one step can take, etc.) the team was able to reduce the hiring process from a maximum of three hundred days to a maximum of seventy-five days.

Looking to the Horizon: Where Are We Headed Next?

Building on our progress in developing a culture of continuous improvement within city government, our next steps are to roll out those same processes into the community. Even as we continue to work to make government better, we recognize that one of government’s most important roles has to do with convening other actors to address community problems that reach far beyond government’s capacity to address alone. To that end, we see the future of Louisville’s Performance Improvement journey as having three parts.

Taking Open Data to the Next Level

• Benchmarking to meaningfully compare Louisville’s performance to other cities or organizations.
• Creating geospatial tools to provide more accurate locational information to make data more useful and relevant.
• Releasing underlying data supporting LouieStat KPIs to allow citizens, developers, etc., to access “raw data” to apply their own analyses and unlock additional insight.
• Instituting a more deliberate feedback loop for allowing outside parties to help “audit” (update/refine) the data and information that government holds.

Leveraging Data to Optimize Internal Government Processes

• Using Predictive Analytics to marshal resources for proactive action by departments.
• Cascading improvement efforts throughout city government, including participation in LouieStat by every department

Convening Community Partners, Using a Common Set of Metrics to Drive Coordination

• Creating issue-based Stat Programs to take the principles of LouieStat beyond department-centric challenges & metrics to include community partners

Louisville’s First Issue-based “Stat” Program

A great example of where we are heading and how we hope to deploy these advances is the pilot of Vacant & Abandoned Properties Statistics (VAPSTAT), our first issue-based stat program. Similar to LouieStat, which evaluates metrics to assess performance and identify opportunities for improvement specific to individual departments, VAPSTAT will analyze progress against key vacant and abandoned property metrics that cross multiple departments and involve multiple community stakeholders.

The community stakeholders will be engaged to determine the most relevant data points (like the number of code enforcement service requests, foreclosures, demolitions, and the amount of liens collected). With this information, the mayor, his senior management team, and key community players will track trend data to assess the impact of current initiatives and identify new tactics or operational changes that must be made to ensure we reach our goals—with the ultimate objective of eradicating vacant and abandoned properties from our community. This information will be presented and discussed in an open forum, where individual citizens, neighborhood groups, and non- and for-profit organizations can bring their data and resources to the table to provide a comprehensive community approach to the complex problem.

VAPSTAT will build on the city’s open data resources in several ways.

Benchmarking Against Peers

VAPSTAT will benchmark Louisville’s data against BlightStatus in New Orleans and other cities with high levels of vacant properties, like Detroit and Philadelphia, to understand the scope of the problem and progress relative to other cities.

Creating Geospatial Tools

VAPSTAT will overlay vacant and abandoned lots/structures with fire, type of crime, and property violations through our code enforcement database, using site addresses to help understand how vacant and abandoned properties correlate with other issues.

Putting Raw Data in Exportable Formats

Our open data portal currently contains datasets for property maintenance, vacant and abandoned property listings, and crime data.

Using Predictive Analytics

VAPSTAT will combine water and electrical connectivity, postal service data, and property maintenance data to anticipate and better plan for resources needed to address problems through proactive outreach and prevention work in more targeted areas.

These tools will enable government and community partners to meaningfully engage in a dialogue that is focused on measurable results, not just words. Effective interventions will be immediately seen and highlighted, enabling the community to replicate what has worked. Most importantly, any organization seeking to get engaged in resolving the issue will have a clear way to engage with other community players—catalyzing partnerships, reducing duplication of efforts, and leveraging existing resources.

Conclusion

Along our journey in Louisville, four major takeaways have helped us get started and build momentum. First, recognize that data will never be perfect—start with the data you have now (in startup speak, a “minimum viable product”) and improve as you go. Second, use data to open a conversation about strategy and prioritization. Third, engage and support your “first followers”—they become your ambassadors and cheerleaders. And finally, celebrate success. It doesn’t have to be about financial rewards; recognition and praise go a long way.

Data for transparency’s sake is a great starting point for driving change. But that’s all it is: a starting point. The trick is to convert the data into useful information that can be used to identify ways to continually improve performance. Once that becomes the focus, then you will be surprised at how many “intra-preneurs” will spring up to help make government better.

With the right support, data enables meaningful conversations about strategy and planning, which, when aligned with the right incentives and skill-building opportunities, can create a culture of continuous improvement. That is the true goal of government reform and the driving mission of our work in Louisville.

About the Authors

A native of Connecticut, Theresa Reno-Weber moved to Louisville in 2010 with her husband, who was born and raised there. She most recently worked with McKinsey & Company, a top-tier management consultant firm that works with businesses and governments across the globe. She has worked with fortune 100 companies, government agencies, and non-profits across the US, Europe, and the Middle East. Theresa is a former Lieutenant in the US Coast Guard, where she served for ten years before earning a Master of Public Policy from Harvard University’s Kennedy School of Government.

Beth Niblock was appointed to be the first CIO for Kentucky’s newly merged Louisville-Jefferson County Metro Government in 2003. She was recognized as one of Government Technology’s Top 25 Doers, Dreamers, and Drivers in 2011.

References

• Collins, J. (2001). Good to Great. New York, NY: HarperBusiness.
• Miller, K. (2006). We Don’t Make Widgets. Washington, DC: Governing Books.
• Ries, E. (2011). The Lean Startup. New York, NY: Crown Business.
• Thousand, J. S. & Villa, R. A. (1995). Managing Complex Change Toward Inclusive Schooling. In R. A. Villa & J. S. Thousand (Eds.) Creating an Inclusive School. Alexandria, VA: Association for Supervision and Curriculum Development.