When selecting the most important government phrase of the last 12 months, “government efficiency” is likely to be a strong contender. But for state and local government officials, it is more than a catch phrase, it’s a way of life. With close proximity to their constituents, these leaders are required to focus on practical efficiency approaches that deliver better impact and value for their communities.
As such, the Center for Digital Government and Governing are cataloging practical, actionable ways to improve efficiency as part of a three-part series. The first article in the series examined budgeting and spending better — how outcome-focused budgeting, faster grants and smarter procurement move resources to what works and get dollars to communities on time. This second article turns to modern tools: customer experience, artificial intelligence, shared services and information technology.
Modern tools are central to government efficiency because they shape both the experience residents have and the work employees do behind the scenes. These tools are the practical methods and platforms that help people complete tasks with fewer steps, less confusion and lower error rates. They allow for service design that is easy to use, automation that removes routine work, shared back-office capabilities that reduce duplication, and secure technology that is reliable. Taken together they can make state, city and county governments more efficient and effective.
Designing Services For People
Barriers. Many residents encounter government through confusing instructions that have too many steps and require accessing numerous unconnected websites, each with their own login. These hurdles add a “time tax” to every interaction. This occurs when people are not sure where to start, cannot see where they are in the process and often abandon applications midway. As a result, staff spend disproportionate time checking status and answering basic questions rather than doing higher-value work. The net effect is slower outcomes for residents and higher costs for agencies.
Solutions. A practical starting point is to map a customer’s journey from start to finish. Journey mapping turns a vague problem — “people find this hard” — into a concrete list of steps, screens, touchpoints and pain points. With that journey map in hand, teams can simplify the route: organize services behind a clear “front door,” use plain language and mobile-ready forms, and monitor cycle times for key items. It often helps to add a simple status tracker so people know where they stand and what comes next, reducing the number of “where is my case?” inquiries that pull staff away from higher-order work. Adding customer-friendly tools, such as those in Texas by Texas, allows for reminders for renewals or automatic notification of missing documents that can prevent cases from stalling.
What to do next. It may help to start with a life experience and map it end-to-end. By focusing on these common requests, governments can deliver high value for residents. If the team has the capacity, explore how a “one-account” model, like New York City’s MyCity portal, could let residents carry their profile and documents across services.
Measuring success. To track progress, leaders can track digital completion rates for the target journey, average time to complete, where drop-offs occur and customer satisfaction. A sustained reduction in case inquiry calls or emails is a strong signal that the new design is working. Setting clear (and public) targets — such as higher completion rates on the most important journeys and meaningful reductions in time to complete — keeps attention on results residents can feel.
Practical Artificial Intelligence That Lifts Capacity
Barriers. Using artificial intelligence seems like an easy way to improve efficiency but many AI initiatives are focused on gimmicks rather than improving outcomes. Even in cases where AI is focused on core business issues, teams may not have clear baselines or targets, so it becomes hard to know whether the tools are helping. Access to the right data is uneven and concerns about privacy, accuracy, bias and records obligations are real and can slow decisions. Even when a tool produces a better email message, that improvement does not automatically translate into more cases closed or faster responses.
Solutions. The most reliable approach for AI is to focus on routine, high-volume work where improvements are measurable. Useful applications of AI include routing requests so the right team sees them first, summarizing case files so specialists spend more time deciding than reading, drafting standard notices in multiple languages, and classifying documents to speed intake. Success tends to come from short, well-scoped pilots with clear productivity and quality targets. For example, Syracuse used a machine learning model to predict likely water main breaks, enabling proactive repairs. In addition, focusing on a simple governance framework helps addresses risk and records while ensuring “human-in-the-loop” review until quality is stable. Creating a small enablement group — policy, security, data, procurement and front-line staff — can help teams select use cases, set up access to data responsibly, build basic quality checks into workflows, and monitor results.
What to do next. A practical first step is to select one or two workflows with strong baselines, such as the time to initial response, the number of cases closed per analyst per week or the size of a specific backlog. It can be useful to run time-boxed pilots — 90 days is often enough to see a signal — and to publish a short monthly scorecard that shows throughput, accuracy checks and user feedback. If a pilot does not move the agreed metrics, it is a good idea to pause it and redirect effort to a stronger candidate. Where tools touch sensitive data, consider a lightweight risk review and clear rules for logging and auditing so teams can scale responsibly.
Measuring success. To ensure that AI is leading to real results on government’s highest priorities, leaders can monitor productivity per full-time equivalent on the targeted workflow, the size of the backlog, the share of issues resolved on the first contact and error rates compared with a human-only baseline. A pattern of steady gains on those measures signals that artificial intelligence is creating real capacity rather than adding a new step or just “saving” staff a certain number of hours on unspecified tasks that may not be related to their highest-priority work.
Shared Services That Reduce Duplication
Barriers. Governments’ internal operations are widely seen as inefficient, time consuming and inconsistent. One reason is that many governments ask each agency to run its own version of common back-office tasks such as paying invoices, managing payroll, processing travel reimbursements, onboarding new employees and handling routine procurement operations. This means that employees, vendors and grantees encounter different processes and portals across agencies. In turn, government leaders lack end-to-end visibility and cannot compare performance. The result is duplication, uneven service levels and less time for program staff to focus on delivery.
Solutions. Shared services can address these problems by moving high-volume, repeatable transactions to a central team with a published service catalog and clear response targets. For example, centralizing vendor e-invoicing gives suppliers one place to submit and monitor invoices, which reduces email back-and-forth and missing paperwork. New York state’s Business Services Center is a strong example of how to centralize back-office functions.
What to do next. Governments may want to begin by mapping their highest-volume transactions and documenting current cycle times and error rates. From there, it often makes sense to move one process at a time — starting with the most standardized work — so lessons carry forward. For example, establishing a single vendor help desk gives suppliers a consistent point of contact and reduces duplicate tickets.
Measuring success. Publishing service targets — such as invoice cycle time from receipt to payment, the on-time payment rate and the volume of inquiries that require a live agent — creates shared expectations and a way to see whether the shift is working. Entities like OhioPays, a single statewide invoicing portal, allow for this type of tracking. When those measures improve — and program teams report fewer back-office distractions — shared services are doing their job.
IT That Enables Delivery
Barriers. Many organizations operate a patchwork of technology applications that do roughly the same job and have aging systems that are prone to outages. Residents and employees face inconsistent sign-in processes across systems, which leads to password resets, lockouts and workarounds. Data sits in silos, so leaders cannot see end-to-end performance. Smaller jurisdictions may not have the staff or tools to manage cyber risks, which raises the risk of interruptions to essential services. Taken together, these IT issues slow delivery and add cost.
Solutions. IT modernization does not have to be a “big bang.” Incremental, outcome-driven steps can make a difference. For example, retiring duplicative applications reduces licensing costs and training overhead. Consolidating platforms where it makes sense — choosing one system for ticketing, one for website content and one for analytics — simplifies support and speeds improvements. When a big system change is necessary, focusing on agile development addresses the problem that makes many leaders reluctant to invest in IT: long and over-budget projects that never deliver the desired outcomes.
What to do next. One practical step is to publish an application list — an inventory of systems that records what each application does, who uses it, what it costs, how it overlaps with other tools and whether it is a candidate to keep, modernize, replace or retire. Setting a goal to replace or decommission a defined portion of lower-value systems each year creates steady progress without disruption.
Measuring success. Leaders can track the number of systems consolidated, the spending associated with those retirements, the average time to restore priority incidents and the share of the workforce that uses multifactor authentication. A steady decline in incidents and faster recovery times are signs that residents will experience fewer interruptions and that agencies can deliver more predictably.
What Success Looks Like
Modern tools and services matter when they change outcomes for real people: Digital completion rates rise, questions are answered on the first contact and permits move on a predictable timeline. On the staff side, time spent on manual re-entry and status checks falls, and more hours go to work that requires judgment and expertise. A small set of measures — service completion rates, average time to initial response, cases closed per week and customer satisfaction — can show whether the changes are sticking and where to improve next. When used correctly, these modern tools translate into day-to-day reliability for staff and value for the communities.
(The Center for Digital Government and Governing are cataloging practical, actionable ways to improve government efficiency. Read part 1 in the series: “Better Budgeting and Spending: Practical Efficiency Ideas for State and Local Governments.” Next up, the final article in this series will turn to operations and stewardship — facilities, permitting, fraud prevention, and hiring.)
Jed Herrmann is a director in the state and local government practice at Guidehouse, where he co-leads the firm’s grants management and program implementation practice. A senior fellow at the Center for Digital Government, he previously worked for state and local governments as well as the White House Office of Management and Budget.
