Introduction

Universities aspire to become more data-informed, yet many organize analytics in ways that make achieving that goal difficult. Analysts distributed across functional units tend to develop reports independently, relying on different definitions, tools, and priorities. Financial pressures, enrollment uncertainty, and the rapid emergence of artificial intelligence (AI) increase the demand for enterprise-wide insight based on well-coordinated analytics. And our systems are only as effective as the quality, governance, and consistency of the institutional data on which they depend.

For all the analytical talent universities employ, limited institutional coordination hinders performance and prevents enterprise-level strategic insight. Many institutions operate in fragmented analytics structures, with analysts scattered across departments, using inconsistent methodologies, and developing skills unevenly. Such decentralization evolves organically, as the analytics function permeates functional units to varying degrees over time. This fractured approach limits the power of critical cross-unit data analysis.

The goal is not simply to produce better reports or centralize more staff, but to build analytics as an institutional capability that continuously integrates data, expertise, and decision-making across organizational boundaries. Like finance or information technology, analytics should be considered a shared infrastructure that enables every division to perform more effectively, independently and collectively. We argue that universities should organize analytics as a hybrid model that centralizes governance, infrastructure, and professional development while keeping analysts embedded in the divisions they support (Horissian & Martin, 2026). This approach strengthens data governance and expands institutional capacity without asking departments to give up the local expertise they rely upon.

A Familiar Organizational Question

Organizations wrestle with the question of centralization versus decentralization, and experience suggests the answer is rarely one extreme or the other. Effective structures balance efficiency and control against adaptability and responsiveness (Daft, 2021a). Centralized structures support standardization and coordination, while decentralized structures support flexibility and domain expertise. For complex organizations like universities, neither extreme serves well on its own. Shared governance prizes collaboration, but unit autonomy often reigns supreme, even as data systems and compliance requirements demand coordinated oversight. As such, contingency theory is often embraced, which posits that structures change only when managers determine that the organization’s structure no longer fits the demands of the situation (Donaldson, 1996). We suggest that the moment may have arrived for institutional analytics.

The Costs of Staying Decentralized

Analytics capacity in many institutions has grown over the past decade, often in a piecemeal fashion inside admissions, finance, advancement, or academic affairs. That disparate growth is understandable, but it carries real costs. Different units build parallel dashboards, maintain separate datasets, or license overlapping software, wasting resources and producing inconsistent outputs. Without shared definitions, the same metric can be calculated differently across offices, leaving leaders without a reliable read on institutional performance.

More importantly, fragmentation keeps institutions from using their analytical capacity as a whole. An institution may employ talented analysts throughout the university and still lack the organizational capacity to combine those individual insights when broader institutional questions demand it.

For example, how do admissions trends relate to financial outcomes, student success, and alumni engagement? All of these data sources are typically housed and analyzed separately and, too often, independently.  

What a Coordinated Analytics Organization Makes Possible

Bringing analytics into a unified office addresses these problems and opens new opportunities.

A holistic institutional view. When enrollment, finance, HR, advancement, student outcomes, and other data can be analyzed together, leadership gains a far clearer picture of institutional dynamics, supporting enrollment planning, resource allocation, and long-term financial modeling (EDUCAUSE Review, 2024a).

Consider a university preparing for its annual enrollment planning cycle. Admissions analysts understand applicant behavior, finance analysts model tuition revenue, financial aid analysts evaluate discounting strategies, and student success analysts identify retention risks. In a fragmented structure in which these analyses occur independently, it is difficult to see how one affects the other.

In a hybrid analytics organization, analysts work together to produce a single institutional forecast that integrates enrollment, financial aid, retention, and net tuition revenue. Rather than serving only enrollment management, this shared forecast informs decisions across the institution, including staffing plans, marketing investments, course scheduling and section caps, housing capacity, orientation planning, and budget development, giving leadership a coordinated, institution-wide view for strategic decision-making. Instead of each office optimizing its own decisions, the institution optimizes the entire enrollment ecosystem.

Stronger governance. Centralization lets an institution set consistent definitions and reporting standards rather than leaving each department to its own practices. Fragmented data ownership raises real risks around compliance, security, and consistency (FormAssembly, 2024)—risks that grow with cloud-based systems and integrated platforms.

AI fundamentally changes the value of coordinated analytics. Predictive models, generative AI assistants, institutional knowledge systems, and intelligent agents all depend on trusted data, consistent definitions, and coordinated governance. Without an enterprise-wide analytics function, institutions risk building multiple AI solutions that answer the same questions differently. Hybrid analytics organizations provide the governance, data standards, and institutional coordination needed to ensure that AI is implemented consistently, securely, and aligned with institutional priorities. Consistent governance also builds confidence among institutional leaders, who can make strategic decisions knowing that the underlying data, definitions, and analytical methods align across the university.

Less duplication. Coordinating tools and workflows enables institutions to build shared dashboards and data pipelines rather than maintaining parallel reporting processes. Centralized structures reduce duplication and improve consistency (Huron Consulting Group, 2016), allowing analysts to spend more time generating strategic insights instead of producing routine reports. As institutions face increasing financial pressure, reducing duplicate reports, technologies, and analytical effort improves efficiency while creating opportunities to redirect scarce resources toward higher-value institutional priorities.

A stronger workforce. A unified team gives analysts room to build skills in statistical modeling, programming, and machine learning. Centralized teams accelerate skill development and scale advanced capabilities more effectively than dispersed teams (Deloitte, 2017). And they also create opportunities for analysts to deepen their expertise, take on more advanced work, and grow professionally while expanding the institution’s analytical capabilities.

The difference is that the institution has the capacity to assemble people, data, and methods around questions that no single division can answer alone. Enterprise analytics is less about organizational ownership than about institutional reach.

The Hybrid Model

Full centralization understandably worries departments, who fear losing responsiveness or working with analysts unfamiliar with their work. A hybrid model addresses this directly; analysts formally belong to a centralized office but remain primary liaisons to the divisions they support, maintaining strong relationships and subject-matter expertise .

A hybrid analytics model distinguishes between capabilities that benefit from enterprise coordination and those that depend on close functional relationships. Data governance, institutional definitions, analytics infrastructure, advanced methodologies, and professional development can be coordinated centrally, while analysts maintain the subject-matter expertise and relationships needed to support individual divisions.

For example, during peak enterprise-level periods such as enrollment forecasting and planning, student success analyses, accreditation reporting, or major strategic initiatives, analysts can be dynamically allocated to high-priority institutional work while maintaining their functional roles. The institution benefits from functional expertise within enterprise-wide coordination, allowing knowledge and best practices to flow across traditional boundaries.

The central office thus provides shared governance, training, infrastructure, and coordination, as well as the flexibility to temporarily redeploy analysts to high-priority projects. The analytics team becomes a shared institutional resource rather than an isolated departmental staff. This also makes them more responsive to faculty and staff who often seek answers to questions that cross departmental lines.

The greatest advantage of a hybrid organization may be that analytical capability compounds over time. A predictive model developed for admissions can produce code, validation procedures, data pipelines, or modeling techniques adapted to student success or advancement. A definition developed for one dashboard can become part of an institutional data standard. A complex analysis completed once can become a repeatable process rather than a one-time product. Over time, the institution accumulates not just reports, but reusable analytical infrastructure and knowledge.

From Service Function to Strategic Partner

When implemented well, a hybrid analytics office moves beyond a reporting service to become a strategic partner in decision-making. By coordinating governance, enabling cross-functional analysis, and surfacing emerging trends, the team can directly inform leadership priorities, shifting from reactive reporting to proactive insight generation that continuously improves institutional performance.

Putting It into Practice

Successful implementation requires deliberate change management, not simply a new organizational chart. Institutions should begin with functions that benefit most from coordination, such as data standards, enterprise reporting, and advanced modeling. Analysts can begin working together on cross-functional projects even before reporting relationships change, creating the foundation for common tools, coordinated processes, and a genuinely unified analytics team. As such, analysts can be aligned under centralized leadership while maintaining their departmental homes.

Framing matters as much as mechanics. Centralization is not a reduction in departmental capabilities, but a strengthening of capacity through shared infrastructure, broader expertise, and coordinated leadership. Institutions that begin with governance and shared standards before changing reporting lines will be more successful than those that focus primarily on organizational restructuring.

Conclusion

For institutional research professionals, this shift also presents an opportunity to lead in analytics governance and the responsible adoption of AI, rather than just supporting it. As complex institutions face growing demands for data-informed decision-making, analytics must evolve from fragmented departmental functions to a coordinated institutional capability. A hybrid structure, one that consolidates governance, infrastructure, and professional development while preserving embedded expertise, offers a practical path, one capable of generating the insights that guide planning, improve efficiency, and support long-term institutional success.

References

Daft, R. L. (2021a). Organization theory and design (13th ed.). Cengage Learning.

Daft, R. L. (2021b). Organizational structure and design. Cengage Learning.

Deloitte. (2017). Analytics operating model. https://www.deloitte.com/ca/en/services/consulting/perspectives/analytics-as-a-living-ecosystem.html

Donaldson, L. (1996). The normal science of structural contingency theory. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (pp. 57–76). Sage Publications.

EDUCAUSE Review. (2024b, August 5). IT centralization patterns and CIO reporting: An update. EDUCAUSE Review. https://er.educause.edu/articles/2024/8/it-centralization-patterns-and-cio-reporting-an-update

EDUCAUSE Review. (2024a, November 18). Data centralization: Enhancing decision-making for transformational change [Sponsored]. EDUCAUSE Review. https://er.educause.edu/articles/sponsored/2024/11/data-centralization-enhancing-decision-making-for-transformational-change

FormAssembly. (2024, November 22). Why data centralization matters for higher education. https://www.formassembly.com/blog/centralized-data-in-higher-education/

Horissian, K. and Martin, E. (2026). “Here is a strong case for hybrid enterprise services.” University Business. September 22, universitybusiness.com.

Huron Consulting Group. (2016). Shared services: Finding the right fit for higher ed. Huron Consulting Group. https://www.huronconsultinggroup.com/insights/shared-services-finding-right-fit-for-higher-ed