This shift can feel overwhelming and unpredictable for city administrators, communications professionals, and technology leaders who might not yet be steeped in AI. But it doesn’t have to be. Understanding the value GenAI and agentic solutions can create — and how to prepare for its unique consumption-based models — will be critical to ensuring your agency remains responsive, efficient, and resident-centric.
"As a former IT manager and performance executive, I know cities are focused on providing the most efficient services to residents. GenAI technologies are capable of doing amazing things that can improve satisfaction and drive down costs to serve. It’s important that government leaders begin to work with finance and procurement — and ideally their data and IT governance teams — to define new processes for budgeting for GenAI."
Kate May
Principal Product Manager, Granicus
The shift: New models of AI procurement
Traditional government technology procurement has relied on predictable, upfront licensing models. You buy software, install it, and pay for maintenance. GenAI flips that model on its head.
The shift in AI-first technology creates new opportunities for IT, city management, legislatures, and other parts of government to explore new budgeting approaches to ensure they can take advantage of these new technologies.
Governments should anticipate their overall investment in AI-enabled technologies is likely to increase significantly over the coming years. To ensure adequate budget — not just for the technology, but to support the strategy, change management, and optimization of these solutions — governments need to explore budget reserves and prioritize investments in centralized agencies to help invest in new skill development, workflow mapping, content strategy and planning, and other key services.
As these new technologies are leveraged effectively — implemented well and aligned with organizational objectives — AI solutions will become budget neutral and even result in cost savings and access to new revenue streams.
Why this matters for government leaders
GenAI isn’t just a tech trend — it’s a new interface for democracy. It can:
Streamline government call volume
Reduce staff burden by automating routine inquiries. Our research with hundreds of governments has found many see up to 60% of their calls to their call centers are for “informational” inquiries — most of which could be self-serviced online. While these calls only last 2-3 minutes, they each incur a cost of $2-$3 for most cities. And, given that most governments only operate call centers between 8 a.m. and 5 p.m., important inquiries from residents after hours go unanswered or are diverted to expensive third-party call centers, staffed by people who are often only trained on a fraction of the knowledge as full-time call center operators.
Serve diverse community needs
Improve equity by making services more accessible to everyone, including non-English speakers or those with disabilities. When these tools are tuned properly, they can ensure responses are provided in plain language so that even guidance on complex topics like regulatory or tax information can be understood by the full community, regardless of their education level or background.
Deliver faster, smarter service
Increase trust and satisfaction by delivering faster, more consistent, and accurate responses. Timely, accurate answers ensure people feel understood and respected — driving satisfaction, trust, and continued digital engagement.
But to unlock these benefits, governments must rethink how they plan, procure, support, and scale AI solutions.
Top 3 tips to prepare for GenAI consumption models
1. Start small, but plan to scale
Begin with a pilot — perhaps a GenAI agent for your city’s website or a virtual assistant for permit applications. Think about what pain points can be solved for your staff and residents by having better access to relevant, personalized information, and ensure you have content available for your agent to generate those most common, useful responses.
Track usage, outcomes, and resident satisfaction. Use this data to model future demand and inform budget forecasts. Always start with a measurable outcome in mind — like a reduction in call volumes or decrease in walk-in traffic — to demonstrate ROI.
Tip: Tip: Ask vendors for tiered pricing models and usage dashboards to help you monitor and manage costs in real time.
2. Build cross-functional AI literacy
Bring together IT, communications, finance, and service delivery teams to build a shared understanding of GenAI’s capabilities and implications. This isn’t just an IT issue; it’s a service transformation.
Tip: Host internal workshops or lunch-and-learns to demystify GenAI and align on goals.
3. Budget for growth, not just deployment
Consumption models mean your costs can rise as adoption increases. That’s a sign of success — but only if you’re prepared. Build flexible budget lines that can expand with usage and consider creating a “digital services reserve” to accommodate spikes.
Tip: Treat GenAI like a utility — budget for it the way you would electricity or cloud storage, with room to grow.
Final thoughts
Generative AI is not a future technology — it’s here, and residents are already expecting it. Governments that embrace this shift will be better positioned to deliver responsive, inclusive, and efficient services.
Understanding and preparing for rapid growth in AI-utilization among your residents is key and is something government leaders should anticipate. ROI should be measurable in months, not years — and it’s important that government leaders don’t panic when resident demand spikes but instead recognize that success and prepare for increased utilization. When this happens, you know you’re achieving your goals — making government accessible for residents 24/7.
At Granicus, we’re committed to helping public sector leaders navigate this transformation with confidence. Whether you’re just starting or scaling up, now is the time to prepare for a new era of digital government — powered by AI, guided by purpose, and built for the people.
