Making better decisions with data in the public sector: A simple guide to modernization
In many cases, the source of these problems can be traced to outmoded ways of organizing, governing, and standardizing the use of data. In many organizations, legacy systems require staff to manually gather data, send files, and approve access for each resource, thereby hindering the use of data. Incomplete data taxonomies and catalogs and a lack of standardization across the organization makes it difficult to use the data to its full potential. These fragmented data sources and manual integration methods hold back new data projects leaving organizations struggling to meet the pace of future needs.
Embracing data modernization in the public sector
To avoid being left behind, public sector organizations are recognizing the need to aggressively embrace data modernization. But what does that mean in practical terms? It involves winding down existing data centers, moving to cloud-based data warehouses and business intelligence platforms, and retraining IT staff to work in other roles. The overarching goal is to take departmental-level legacy data sources and apply governance to make them consistent, and to reduce errors by cleansing, standardizing, and integrating data into a cloud data warehouse to create a single trusted view of information. Once the data is centrally located, your organization can apply business analytics to it and gain the business awareness inherent in it.
To avoid being left behind, public sector organizations are recognizing the need to aggressively embrace data modernization.
The result? Improved cross-functional collaboration, efficiency, streamlined regulatory compliance, and a robust platform built for whatever the future can throw at it. In short, it’s all the necessary items to enhance understanding of workflows for better, faster decision-making and improved outcomes.
A public sector data modernization blueprint for success
A typical data modernization blueprint begins with an Analytics Center of Excellence (ACE) — a cross-functional team of stakeholders focused on identifying needs, defining data quality standards and compliance procedures, and identifying, prioritizing, and coordinating a portfolio of data analytics projects in accordance with the organization’s strategic priorities. This results in two important outcomes: data democratization — the ability to access data at the enterprise level, and data literacy — the ability for people in the organization to understand what data is available to help make data-driven decisions.
A typical data modernization blueprint begins with an Analytics Center of Excellence.
The ACE can be built from the ground up or built on what’s currently in place. If you have a preexisting data warehouse or the beginning of one, the first step is to look at the various technologies and determine whether they can be scaled to an enterprise level. Next, and often the bigger factor, is setting up data governance. Your organization may have the technology in place to build out warehouses and integrations but lack maturity on governing the assets that are being created. This includes building a structure to support the literacy — for example, business glossaries, lineage, and cataloging — and understanding the various roles and the responsibilities each role should have.
From there, a typical project proceeds following these steps:
1. Assess the current environment
Start your data modernization effort with a deep assessment of the current environment to understand your existing technologies and skills, the processes that are being done manually, how your end users are being supported, and the trust level in the data they currently have. Ask these questions:
- Do you have the right analytics tools, enabling technologies, and infrastructure to realize the value of your data and meet the strategic goals of your organization?
- Does your organization consistently use efficient, best practice analytics processes?
- Does your staff have the right skill sets to optimize available data and quickly provide you with actionable insights?
Your assessment should extend beyond the stakeholders to include end users since there’s frequently a disconnect between stakeholder perception and the reality on the ground.
2. Clarify stakeholder vision
Next, the team takes the information gained during the assessment and collaborates with all stakeholders to define a vision of what they want the data modernization effort to become. Is there a distinct strategic vision? Or is it something more broad, like, “We need to do better.” Then craft a vision statement.
3. Create a roadmap and execute
From there, you’ll develop a project roadmap based on available budget, resources, and skill sets. What your roadmap looks like will depend on what the current state looks like and what your priorities are at that given time. The team will review the assessment to see what what’s being done now and decide what’s being done well and what gaps should be filled to add the most value. Many times, governance needs to be tackled first. In other cases, existing technology is making integrations difficult — an indication that it needs to be upgraded. This will require the roadmap to initially focus on technology in order to streamline later development and quicken timelines to meet priorities.
The plan can be anywhere from six months to five years depending on your organization’s goals. If a larger budget is available, it may be possible to tackle a greater amount in a shorter period of time. If the budget isn’t as broad, or resources don’t allow for an “all-at-once” approach, extend out the timeline, identify the priority pieces based on the vision and current environment, and start working toward incremental improvements. The ways into it are as varied as the organizations adopting it.
A successful project starts at the top. Executive sponsorship is critical to keep people moving in the same direction, ensure buy-in, and make sure staff make the necessary changes to their processes. In the most successful projects, leaders recognize the opportunity, clearly understand it will take several years firing on all cylinders to make it successful and become closely involved in seeing it through.
Other factors include:
- Team selection: Having the right team can make or break a project. Identifying the right personas on the operational and business side is critical. Find out what’s important to each team member, their use cases, the problems they currently have with data, and how things can be improved.
- Education of stakeholders: Explain to stakeholders what’s happening at each stage, and plan for small, quick wins so everyone sees continuous progress. The most successful projects are phased over time rather than trying to do everything all at once. This enables you to prepare people for change, show them value at each phase of the project, and bring them along as active participants.
- Governance: Governance and the concept of ACE is important to establish how everything intersects and relates. The ACE needs to be very robust and include technology governance, all data delivery mechanisms, and what will be done with the legacy data.
- Communication: “Over communication” at the executive level is a key factor in successful projects. Follow up with operational and business collaborators at regular intervals — say monthly — to show them what you’re doing to address their problems. Remember to show them the quick wins to demonstrate progress and commitment.
Data management is a long-term journey. Regardless of how much you’re able to tackle at the onset, the bottom line is it’s always evolving, whether it’s enhancements and new things you can learn from your data or keeping up with constantly changing laws and regulations. At the end of the day, that’s what an analytics environment is ultimately for — to help your organization advance along with society and the business you’re supporting.
Data management is a long-term journey. Regardless of how much you’re able to tackle at the onset, the bottom line is it’s always evolving.