BI Reporting, Data Warehouse Systems, and Beyond (240823865)

Published on December 2016 | Categories: Documents | Downloads: 11 | Comments: 0 | Views: 102
of 11
Download PDF   Embed   Report

This Spotlight focuses on data from the 2013 Core Data Service to better understand how higher education institutions approach business intelligence (BI) reporting and data warehouse systems (see the Sidebar for definitions). Information provided for this Spotlight was derived from Module 8 of CDS, which contains several questions regarding information systems and applications. Responses from 525 institutions were analyzed. Only U.S. institutions were analyzed for this report. In addition, we interviewed four subject-matter experts to gain insights about the current and future state of BI reporting and data warehouse systems in higher education.The ECAR CDS Spotlight Research Bulletin Series highlights findings from the EDUCAUSE Core Data Service, focusing on a small but meaningful slice of data collected in the CDS. These selected highlights are intended to provide context and meaning for CDS benchmarks that may be of especially broad interest, be especially timely, or draw connections between research from ECAR and CDS. The series is featured along with other CDS publications on the CDS website, http://www.educause.edu/coredata, and is now available to eligible ECAR subscribers as part of the ECAR subscription.Citation for this Work: Leah Lang and Judith A. Pirani, BI Reporting, Data Warehouse Systems, and Beyond, research bulletin (Louisville, CO: ECAR, April 23, 2014), available from http://www.educause.edu/ecar. ECAR research helps you predict, plan for, and act on IT trends in higher education. Subscribe now. http://www.educause.edu/library/resources/bi-reporting-data-warehouse-systems-and-beyond

Comments

Content




1
Research Bulletin Apri l 23, 2014
BI Reporting, Data Warehouse
Systems, and Beyond
CDS Spot l i ght Report
© 2014 EDUCAUSE and Leah Lang and Judith A. Pirani.
CC by-nc-nd.
Leah Lang, Manager, Core Data Service, EDUCAUSE
Judith A. Pirani, Consultant, EDUCAUSE, and President, Sheep Pond Associates
Overview
This ECAR research bulletin series highlights findings from the EDUCAUSE Core Data
Service, focusing on a small but meaningful slice of data collected in CDS. These selected
highlights are intended to provide context and meaning for CDS benchmarks that may be of
especially broad interest, be especially timely, or draw connections between research from
ECAR and CDS. The series is featured along with other CDS publications on the CDS
website at http://www.educause.edu/coredata and is now available to eligible ECAR
subscribers as part of their subscription.
This Spotlight focuses on data from the 2013 Core Data Service to better understand how higher
education institutions approach business intelligence (BI) reporting and data warehouse systems
(see the sidebar for definitions). Information provided for this Spotlight was derived from Module 8
of CDS, which contains several questions regarding information systems and applications.
Responses from 525 institutions were analyzed. Only U.S. institutions were analyzed for this
report. In addition, we interviewed four subject-matter experts to gain insights about the current
and future state of BI reporting and data warehouse systems in higher education.
BI Reporting and Data Warehouse Systems:
Now and Beyond
Data collection, management, and analysis continue to grow more important in higher
education in response to issues such as regulatory and accreditation requirements, student
success measurement, and operational efficiency expectations. In addition, parent and student
use of personal online banking and shopping applications fuel similar capability and data-
transparency expectations in areas such as tuition payments and academic performance
evaluation.
These pressures promote an ongoing cultural shift in higher education, one that embodies
greater measurement, assessment, and accountability. With this change comes an increasing
recognition of the tactical need and strategic value of using institutional data. BI reporting and
data warehouse systems are two mainstays in an expanding portfolio of tools and capabilities
that enable senior leaders and others to unlock the inherent value of institutional data to
address compliance and regulatory needs, to enhance users’ experience, and to monitor
progress on short-term goals and long-term strategies. Perhaps this is why these systems are
the fifth (BI reporting) and sixth (data warehouse) most rapidly changing core system areas
(see figure 1).
1


2
Figure 1. Characteristics of Core Information Systems

Current Snapshot
Institutions with BI reporting and/or data warehouse systems are poised to innovate in their use
of institutional data. According to CDS, in FY 2012/13, BI reporting and data warehouse
systems were relatively common at U.S. institutions (see figure 2). Approximately 80% of U.S.
institutions had some type of BI reporting system, and 71% of U.S. institutions had some type
of data warehouse system; both system types are more pervasive at larger schools than
smaller schools. The presence of these systems has increased in the past three years, and it
appears this growth will continue in the next three years. CDS data from FY 2010/11 show that
only 62% of institutions had a data warehouse/BI system. In the next three years, an additional
8% of institutions intend to implement a BI reporting system, and an additional 12% intend to
implement a data warehouse system.

3
Figure 2. Percentage of U.S. Institutions with Systems Implemented or Planned
Implementations

As institutions implement new systems, the debate over centralized versus decentralized
management will continue. Most institutions with existing systems rely on centrally managed
systems to enable better data management in a number of ways: a central extract, transform,
and load (ETL) layer for information access; a common toolset for reporting; and “de-siloed”
information for common data sets. About 62% of U.S. institutions have BI reporting systems
provided by central IT; 59% of U.S. institutions have data warehouse systems provided by
central IT. The theme of central control is echoed with an overwhelming preference for in-house
management of both BI reporting (90% of U.S. institutions with a BI reporting system) and data
warehouse systems (92% of U.S. institutions with a data warehouse system) over vendor or
cloud-based solutions. In fact, these systems are the least likely to be outsourced out of our list
of core systems (see figure 1; BI reporting outsourcing ranking = 19, data warehouse
outsourcing ranking = 20).

4

Institutions interested in building analytical capabilities have more to consider than simply
implementing systems and deciding where they should be managed. A robust analytic
environment also consists of integrating these systems in a way that produces meaningful and
actionable analyses. Although many institutions have BI reporting and data warehouse
systems, the 2014 ECAR report on higher education’s top-ten strategic technologies suggests
that few institutions are configured in a way that supports analytic capacity (20% of institutions
have BI reporting dashboards in place for analytics; 35% of institutions have a data warehouse
in place for analytics).
2
As institutions prioritize analytics, however, efforts are being made to
integrate these systems and prepare them for this purpose (59% of institutions are investing in
planning and implementation of BI reporting dashboards for analytics; 43% are investing in
planning and implementation of data warehouse systems for analytics). These institutions may
plan to implement new systems and/or leverage existing systems. A variety of analytic
environments could emerge, given the number of current combinations of solution types in
place. The most common combination of systems among institutions with either a BI reporting
or data warehouse system was a data warehouse solution from an infrastructure provider with a
single BI reporting solution (21%; see Table 1).

CDS Spotlight Definitions
ad hoc query: A nonstandard inquiry to obtain information as the need arises. This contrasts with a query
that is predefined and routinely processed.
analytics: The use of data and statistical analysis and explanatory and predictive models to gain insight into
and act on complex issues.
big data: Massive amounts of data, collected over time, that are difficult to analyze and handle using common
database management tools. Big data can include business transactions, e-mail messages, photos,
surveillance videos, and activity logs.
business intelligence reporting: A set of administrative functions and associated software systems that
support planning and decision making by categorizing, aggregating, analyzing, and reporting on data resulting
from transaction-processing systems.
data warehouse: A central repository of data often created by integrating other data sources and used for
reporting and analysis.
extract, transform, and load (ETL): The functions performed when pulling data out of one database and
placing it into another of a different type. ETL is used to migrate data, often from relational databases into
decision-support systems.
Hadoop: An open-source project from the Apache Software Foundation that provides a software framework
for distributing applications on clusters of servers; designed to handle huge amounts of data.
predictive analysis: Methods used to extract information from data and use it to predict trends and behavior
patterns.
Sources: CDS Glossary, EDUCAUSE, PCMag.com Encyclopedia, and Wikipedia.

5
Table 1. Variety of BI Reporting and Data Warehouse Systems Reported
BI Reporting
Solution Type
Data Warehouse Solution Type
Row
Total* Homegrown
BI/Infrastructure
Provider
Admin.
Systems
Provider None
Same
Solution as BI
Reporting
Solution Other
BI Provider
(single)
14% 21% 12% 11% 5% 0% 63%
Homegrown 10% 1% 1% 3% 0% 0% 15%
BI Providers
(multiple)
1% 4% 0% 0% 1% 0% 7%
Admin. Systems
Provider
1% 1% 3% 1% 0% 0% 6%
None 3% 1% 1% 0% 0% 0% 6%
Admin. Provider
with Third-Party
BI Tools
0% 0% 1% 1% 0% 0% 2%
Other 0% 0% 0% 0% 0% 0% 1%
Column Total* 29% 28% 19% 17% 6% 1% 100%
* Certain totals differ from apparent sums due to rounding.

Even with the range of possible of solutions, customization continues to be an avenue for fine-
tuning system configurations to match business processes. Approximately 30% of U.S.
institutions have substantially customized BI reporting systems (ranked eleventh in figure 1).
Data warehouse systems tend to require more customization than BI reporting systems (46% of
U.S. institutions have substantially customized data warehouse systems; ranked third in figure
1). Substantial customization is more common when the data warehouse system is provided
outside central IT (66% of distributed systems versus 42% of centrally provided systems).
Additionally, older systems tend to be more frequently customized. On average, for each
additional year of age of a data warehouse system, the likelihood of substantial customization
increases 7.5%. There is not a significant association between age of BI reporting system and
customization.
As systems age, institutions begin to consider the trade-offs between replacement and further
customization. The BI reporting and data warehouse system areas are relatively young,
however, and few institutions plan to replace either their BI reporting system (13% of U.S.
institutions) or their data warehouse system (14% of U.S. institutions) in the next three years.
Those who are planning to replace systems cite upgraded functionality as the primary reason
(84% of institutions replacing a BI reporting system; 71% of institutions replacing a data
warehouse system).
Finally, unlike the homogenous learning management system market space, in which 93% of
the market uses one of the top-five vendors, the BI reporting and data warehouse landscape is
relatively diverse. Only 52% of the market is using the top-five BI reporting vendors, and only
50% of the market is using the top-five data warehouse vendors. Figures 3 and 4 depict the
market share for the most common BI reporting and data warehouse solutions at CDS
participating institutions.

6
Figure 3. BI Reporting System Market Share
Figure 4. Data Warehouse System Market Share

7
Future Trends
BI reporting dashboards that visually monitor and display the status of processes and activities
in the form of charts or metrics captured the top spot on the list of higher education’s top-ten
strategic technologies for 2014. The list’s sixth-ranked strategic technology was administrative
or business performance analytics to help target organizational resources to support
organizational goals.
3
Institutions characterizing themselves as early adopters of technology will
spend considerably more time than other institutions on five technologies; of these, three relate
to the analytical portfolio in the form of emerging analytical techniques and data sources:
predictive analysis, Hadoop, and big data.
4
As focus shifts to this application area, the following
trends will impact the future of BI reporting and data warehouse systems in higher education:
 Enhancements to the Current BI Reporting/Data Warehouse Platform: Support for
growing data-exploration and analysis requirements may necessitate enhancements to
the current BI reporting/data warehouse platform—for example, greater data
interoperability to create more possible data combinations (e.g., combining HR data with
financial data or housing data with financial aid data) and the use of metadata to promote
consistent understanding of data sources and definition among all data users, not just the
data owners (e.g., metadata for GPA definitions to help users not employed in the
registrar’s office). In addition, the proliferation of smartphones and tablets feeds user
expectations for anytime, anywhere data access, requiring mobile reporting and data-
access capabilities (e.g., an alumni affairs associate downloading a prospective donor’s
gift-giving history before a fundraiser).
 Emergence of the Analytical Portfolio: Users’ growing analytical expertise feeds the
exploration of more complex questions and issues, which eventually may require
dynamic and flexible blending of data as well as cutting-edge data analysis and
visualization tools. Traditional BI reporting tools and data warehouses may not be robust
enough to provide these capabilities, leading to the development of an expanding
institutional portfolio of analytical tools and capabilities to address diverse information
needs. For example, mature data warehouses and advanced BI systems can provide
real-time and accessible reporting, dashboards, and data visualizations; systems that
provide just-in-time advice and alerts can enable students and their advisors and
instructors to take action to improve performance or enable administrators to optimize
services and processes.
5
The deployment of analytics opens the door to the adoption of
new supporting tools and platforms, as for example, predictive analysis.
6

 New Data Sources: Institutions will likely mix new data sources into their analysis. Until
now, the major type of institutional data has been structured data, such as is found in
administrative and academic systems. In the future, higher education will most likely
move to follow the commercial sector’s use of unstructured data (e.g., social media such
as Facebook and Twitter) to respond to customers’ evolving requirements. For example,
GPS and other geospatial data such as radio frequency ID, near-field communications,
and Bluetooth low-energy proximity sensing can provide information on ways to enhance
student services, though privacy issues may curtail some uses of unstructured data.

8
Tips for Successful Implementations
As the list of top-ten strategic technologies for 2014 illustrates, IT professionals recognize the
importance of supporting higher education’s burgeoning analytical culture. Whether IT
organizations are building their first BI reporting tools and data warehouse systems or
expanding current capabilities into an analytical portfolio, our research yielded several relevant
recommendations to consider.
Assess Current Capabilities
Before moving forward, it is advantageous to understand the institution’s current state in order to
focus planning and investment optimally. Maturity models, such as the Oficina de Cooperación
Universitaria BI Model
7
or ECAR’s Analytics Maturity Model,
8
offer institutions a starting point to
determine current strengths and weaknesses and guidance on how to proceed accordingly.
Create a Roadmap
Supporting an institution’s analytical culture is not a project but an ongoing program, and the
program’s direction and approach should be driven by the institution’s aims and questions to be
answered. This requires a strategic plan that provides the big-picture aims, funding, and
resources, as well as a detailed roadmap of the program’s nuts and bolts—how the
implementation will address the information needs of each constituency, as well as details on
training and education in using the tools and understanding the data.
Create a Collaborative, Institutional Effort
There may be a tendency to focus on technology aspects because they are most familiar and
tangible, but people and process are perhaps more important than the underlying infrastructure.
Building analytical capability involves relationships, problem solving, prioritization, and
governance of which problems to tackle—as well as an understanding of institutional culture.
Successful collaborations require compelling need, a dedicated leader, and committed coalition
members.
9
Focusing on the following elements will enable central IT to successfully navigate
an institutional effort to build analytical capability:
 Sponsorship: Senior-level sponsorship is a vital component in any major IT initiative, but
perhaps it is even more critical when supporting analytical capabilities, given their
complexity and expense, as well as their direct impact on senior-leadership functions.
The choice of institutional sponsor is equally critical; one needs a champion who
understands thoroughly the ins and outs as well as the power of analytical tools, so that
person can convey knowledgeable enthusiasm for the program’s possibilities and
benefits to colleagues and to the institution at large.
 Balance between Central and Local: Central IT may manage BI reporting and data
warehouse systems at many institutions, but this approach should not preclude local
involvement. The centralized unit may manage much of the infrastructure, data, and
tools, but some analytical activities will address local (e.g., college), specific needs that
may require the use of supplemental tools and a blend of local and institutionally certified
data. The central and local units must work together to identify local needs and
partnership opportunities.

9
 Governance: Coordinating these central and local interests requires strong governance,
especially in more decentralized institutions. A representative group can prioritize
program activities, but an institution can reap greater benefits from creating a higher-level
steering committee that comprises leaders of the institution’s data-producing and data-
consuming areas. This helps align analytic activities with institutional strategic objectives.
In addition, this steering committee can serve as a core decision-making body that guides
future institutional investment and expansion of analytic capabilities.
 Communication: Grasping the potential and operation of analytic tools can be difficult,
and thus change management and education are very important. An ongoing
communication campaign, commencing at the program’s inception, is required to
demonstrate its capabilities and possibilities to staff members at every relevant level and
area. The message has to be consistent and incessant to promote widespread
awareness and understanding.
Prototype First
As with many complex IT projects, it may be behoove institutions to create a small prototype to
show the value of the analytic solution and expand its capabilities over time. Such a strategy
gives the program sponsor a quick win and establishes trust in future endeavors as the
institution sees an operational example that demonstrates the possibilities and benefits.
Plan for the People
An implementation should take into account individuals in all roles—service owners, operators,
end users, and others:
 Staffing: Staffing has both operational and strategic implications. There is a need for
localized data experts or IT professionals with both functional knowledge and analytical
skill to provide hands-on support to the dean or administrator who requires the
information. For example, each college at Purdue University has a staff member focused
on information, data, and analysis. However, one commonly expressed problem is the
ongoing shortage of qualified individuals to fill these roles, and one potential solution is to
draw on an institution’s most valuable asset—its students—either in the form of
internships or career-development opportunities. To support the strategic use of data and
analysis, some institutions have hired a chief data officer to bring high-level leadership to
these two areas and to coalesce central and local data stakeholders and activities into an
enterprise endeavor.
 Ease of Use: Given the shortage of information-analysis professionals in higher
education, it is imperative to fold ease of use into the program design. One means is to
create common enterprise data sets whenever possible that are available for automated,
institutional applications or that can be combined with local, customized applications.
Again, this approach ties back to maintaining the central/local balance and grooming
partnerships. In addition, some applications and tools, especially in the areas of data
visualization and data discovery, can be complicated for the data novice to use. When
creating or purchasing off-the-shelf tools, keep in mind the skill level of business or
functional staff members who will be using them.

10
Look to Peers
The peer benchmarking data included in CDS can help inform decision making throughout an
implementation effort. Knowing which solutions peer institutions are using, as well as the extent
to which their systems are customized and whether they plan to replace them, can provide
additional insight for planning and roadmaps. Collaborating with peers through networks such
as the Higher Education Data Warehousing Forum
10
and shared services can help uncover
best practices and opportunities for operational efficiency.
Where to Learn More
 Chapple, Michael. “Effective Data Governance Practices.” Presentation at the 2013
Security Professionals Conference, St. Louis, Missouri, April 17, 2013.
http://www.educause.edu/events/security-professionals-conference/2013/2013/effective-
data-governance-practices.
 Collins, Marcus. 2014 Planning Guide for Data Management. Research report. Stamford,
CT: Gartner, December 9, 2013. http://www.educause.edu/library/resources/2014-
planning-guide-data-management.
 Higher Education Data Warehousing Forum. http://www.hedw.org/.
 Lang, Leah. 2013 CDS Executive Summary Report. Research report. Louisville, CO:
EDUCAUSE, February 2014. Available from http://www.educause.edu/coredata.
 Grajek, Susan. Higher Education’s Top-Ten Strategic Technologies in 2014. Research
report. Louisville, CO: ECAR, February 2014. Available from http://www.educause.edu/
ecar.
 Grajek, Susan, Leah Lang, and David Trevvett. The 2011 Enterprise Application Market
in Higher Education. Research report. Louisville, CO: ECAR, June 2012. Available from
http://www.educause.edu/ecar.
 HBR Blog Network, “Are You Ready for a Chief Data Officer?” blog entry by Thomas C.
Redman, October 30, 2013. http://blogs.hbr.org/2013/10/are-you-ready-for-a-chief-data-
officer/.
Acknowledgments
The authors thank Theodore M. Bross, Associate Director, Administrative Systems, Princeton
University; Ora Fish, Executive Director of the Data Warehouse and Business Intelligence, New
York University; Beth Ladd, Director, Business Intelligence, Illinois State University; and Aaron
Walz, Director, Business Intelligence Competency Center, Purdue University, for sharing their
professional experiences and insights about business intelligence and data warehouses in
higher education applications.
About the Authors
Leah Lang ([email protected]) is Manager of the Core Data Service at EDUCAUSE. Judith
A. Pirani ([email protected]) is an EDUCAUSE consultant and President of Sheep Pond
Associates.

11
Citation for This Work
Lang, Leah, and Judith A. Pirani. BI Reporting, Data Warehouse Systems, and Beyond.
Research bulletin. Louisville, CO: ECAR, April 23, 2014. Available from
http://www.educause.edu/ecar.

Notes
1. Rankings for figure 1 are out of 20 system areas covered in Module 8 of CDS 2013. Rate-of-change ranking is
based on age of system, number of institutions implementing the system, and number of institutions replacing the
system. This ranking is an indication of how ripe the market is for vendors in this space (1 = high change; 20 = low
change).
2. Susan Grajek, Higher Education’s Top-Ten Strategic Technologies in 2014, research report (Louisville, CO:
ECAR, February 2014), available from http://www.educause.edu/ecar.
3. The survey was distributed to 10,393 EDUCAUSE members as part of the Top-Ten IT Issues Survey in November
2013; 444 members responded and indicated, for each technology, the attention their institution was planning to
devote to 78 technologies in 2014. The survey defined strategic technologies as “relatively new technologies
institutions will be spending the most time implementing, planning, and tracking in 2014.” For more information,
see Grajek, Higher Education’s Top-Ten.
4. Ibid., 19.
5. Susan Grajek and the 2012–2013 IT Issues Panel, “Top-Ten IT Issues, 2013: Welcome to the Connected Age,”
EDUCAUSE Review 48, no. 3 (May/June 2013), http://www.educause.edu/ero/article/top-ten-it-issues-2013-
welcome-connected-age.
6. For more information about analytics, see Jacqueline Bichsel, Analytics in Higher Education: Benefits, Barriers,
Progress, and Recommendations, research report (Louisville, CO: ECAR, August 2012), available from
http://www.educause.edu/ecar.
7. OCU BI Maturity Model, Jisc infoNet, http://www.jiscinfonet.ac.uk/infokits/business-intelligence/measuring-
success/ocu-maturity-model/.
8. ECAR, “Analytics Maturity Index,” https://www.surveygizmo.com/s3/1020013/Analytics-Maturity-Index.
9. Jack McCredie and Judith A. Pirani, A Dozen Gurus Describe IT Collaborations That Work, research bulletin
(Louisville, CO: ECAR, May 22, 2012), available from http://www.educause.edu/ecar.
10. See http://www.hedw.org/.

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close