That’s because the capabilities of technologies designed for consumer use — email and file storage solutions — quickly surpassed those of legacy enterprise solutions. Driven by free or low-cost pricing, they were first adopted at scale by individuals for personal use —including on their computers at work, which helped convince enterprise technology leaders to take an if-you-can’t-beat-them-join-them approach to adopting them for the workplace. (Dropbox’s inroads into the enterprise using these tactics is the focus of an often cited Harvard Business Review case study).
In many cases, state and local governments followed suit, with their first cloud implementations involving low-hanging fruit like email. But the roadmap for further cloud adoption in the public sector has been less clear.
As government efforts to modernize aging systems take on new importance in the wake of the pandemic, technology leaders face a more challenging task: identifying which applications and use cases make sense to migrate first.
“Don’t try to do everything at once,” urges Center for Digital Government (CDG) Vice President Phil Bertolini, who in 2010 began cloud migration efforts in Oakland County, Mich., where he served as CIO. “There’s a thought of just forklifting everything into a cloud environment, and it doesn’t work that way.”
Top Priorities: GIS, ERP and Application Development
During the pandemic, state and local governments leveraged the cloud’s flexibility to quickly stand up new services, largely around public health solutions addressing the need for information about COVID-19 as well as to coordinate testing and vaccine distribution. Disaster recovery also has grown in importance as governments rethink resilience. Cloud providers offer the ability to host backups and services in geographically disparate locations, helping address continuity of operations in the face of a widespread natural disaster that might impact both traditional data centers and nearby backup sites.
But even before the pandemic, state, county and city governments had consistently prioritized several key systems for cloud migration. Among them are geospatial services such as geographic information systems (GIS), particularly for state and local governments who rely on the technology for a wide range of functions and can tap into mature cloud implementations.
“Why do all that on premises when someone has already invented the wheel?” Bertolini says.
Other top priorities for cloud migration include application development and testing, such as sandboxed cloud environments that help governments meet the need to test, stand up and iterate applications more quickly. This is particularly true in cloud-based low- or no-code environments, in which solutions can be rapidly developed and deployed — even across government entities. Wyoming took this route with its modernized unemployment system, which was designed to be portable so it could be adopted by other states.
Finance and administration and human resources applications are also at the top of priority lists — either as standalone systems or as part of broader enterprise resource planning (ERP) solutions that can incorporate procurement and asset tracking capabilities. These tend to be among the largest and most mission-critical government systems, typically used by many employees across multiple departments, agencies and functions. For that reason, they are ideal for cloud deployments.
Applications addressing the environment, public safety and transportation are among other cloud priorities cited by state and local government leaders in CDG’s most recent digital states and counties surveys.
County Cloud Priorities:
State Cloud Priorities:
Among the strategies for government leaders mapping their cloud migration:
Start simple. Many state and local governments have kicked off cloud migration efforts by identifying relatively small and self-contained applications. “Do that first so you understand the security and migration issues,” Bertolini says.
At the same time, that doesn’t mean that every small application is suited for cloud migration. “Some small packages run on one server and don’t need a lot of care and feeding — government has a lot of these things,” Bertolini says. “You don’t know if they’re going to run better in the cloud or if it’s going to be cost effective.”
Evaluate key operational factors. Among them:
• Latency: Are there issues with connectivity that make certain applications — and particularly mission-critical ones — not as usable as if they were housed on premises?
• Integration: Can a cloud solution integrate with on-premises systems or the data stores they need to draw from? For example, legacy mainframe systems may lack modern application programming interfaces (APIs) for modern data interchange.
• Security: Moving applications or systems to the cloud does not alleviate government responsibility for security. Still, cloud providers have worked to address concerns by providing comprehensive security solutions that few individual organizations can match. Government technology leaders should “supplement that with practices on premises,” Bertolini says.
• Compliance: Governments must ensure cloud solutions are certified to comply with laws and regulations governing the security and privacy of certain types of data — Health Insurance Portability and Accountability Act (HIPAA) compliance for sensitive health information and Criminal Justice Information Services (CJIS) compliance for a range of information used by public safety and justice departments, to give just two examples. These issues are one reason why public safety organizations have been slower to move to the cloud, according to Bertolini.
• Skillsets: If governments don’t have the in-house skillsets to maintain cloud applications, it’s important they determine whether technology providers or other partners can provide them.
• Finances: While there may be immediate cost savings from shuttering systems, it is important to look at long-term ROI — a process complicated by the shift from capital to operating expenditures when applications move to as-a-service models. “There are nuances to managing this from a financial and priority perspective,” Bertolini says.
Celebrate Quick Wins
The end goal of starting small by identifying discrete use cases is to get quick wins which can help provide an impetus for further modernization. Move quickly and document the impact, Bertolini urges. “Once you get that first win, you can demonstrate the benefits to leadership and push your team to go forward,” he says.
Special Content: This article was prepared by the Government Technology Content Studio, which is editorially independent of both the sponsor and the Government Technology editorial staff, who were not involved in producing it.