Manage Learn to apply best practices and optimize your operations.

Create a manageable business intelligence strategy, governance plan

Planning for a successful business intelligence strategy, however necessary, can seem daunting -- especially in the midmarket. Learn how one CIO tackled the early stages of BI.

The benefits of a business intelligence (BI) strategy -- tools, processes and systems that assist in decision making, operations and strategic planning -- are vast. A company's data can be a gold mine of information, helping an organization boost operational productivity and efficiency, increase its understanding of the competitive landscape and find new ways to add to the bottom line.

Scott Lowe
Scott Lowe

At Westminster College, we're in the very early stages of building a business intelligence strategy. Believe it or not, colleges and universities have a real revenue motive around the use of business intelligence tools. Although most private colleges and universities are not for profit, they're also not for loss, and it's critical that we maximize any and all resources we have available.

Colleges, much like our corporate brethren, operate against a number of metrics -- key performance indicators (KPIs) that ultimately determine our success both in the classroom and on the balance sheet. Most recently, we've discovered how important it is to understand our operations at every level. With the right data we can make better decisions both tactically and strategically, and we can hold people accountable against measurable performance goals.

We've created an initial BI governance group to keep efforts at some kind of reasonable initial scope, help evaluate software and be a clearinghouse for all cross-cutting reporting/metric needs. Although we're still early on in our business intelligence strategy efforts, our ultimate goals are lofty and include:

Seamless BI reporting: One-off reporting is great to get a quick answer, but if a report is needed more than once, its creation should be formalized and normalized.

Common reporting: We use a variety of reporting tools and get regular requests to add new ones. We need to establish a BI tool that can be used widely to ensure consistency and accessibility.

Predictive analytics: Our data is a treasure trove of information that can lead us to clues about the future. We need to mine this data to help determine the direction our key metrics are heading so we can adapt appropriately.

We've faced some serious challenges, however, when it comes to using the data. For example, we have data silos operating independently of one another and integrated manually, which results in an unintentional lack of consensus on definitions and, as a consequence, different answers to some basic questions, such as the exact number of students we have on campus.

Very recently, we established a representative business intelligence committee composed of members from each functional division, including finance, IT, enrollment, academics, student life and advancement. As a CIO with significant process enhancement responsibility, I'm chairing this new committee and plan to do the following:

Create a data dictionary. We need a common set of definitions to use for all metrics and KPIs. We're calling this our campus data dictionary. Where it's necessary to use the same definition for multiple metrics (e.g., "How many students do we have?" means different things to the registrar and to campus housing), make sure that appropriate context is provided.

Inventory current efforts. Different campus groups already capture a wide variety of data elements that could be used in a broader, more strategic way. One charge of the BI committee is to inventory all existing metrics, validate their underlying methodology and determine if any of them can be tweaked to be of broader use.

Define new key performance indicators. We'll begin to define and prioritize the KPIs and metrics on which we will initially focus our efforts. We'll also agree upon and validate the methodology upon which the KPI will be based.

Validate data structures. We'll determine what potentially missing data structures and processes are necessary to support capturing the aforementioned KPIs. For example, if we want to understand on a day-to-day basis how many students we have on campus, do we have the appropriate tracking fields in our database? If we don't, what fields are necessary to track the KPI? Kick any new requirements over to two other campus committees charged with implementing new processes and data structures.

Define the data warehouse. We'll create the definition for a data warehouse to store necessary information for trending/historical and, eventually, predictive purposes.

Dashboards/KPI sharing. For some of our efforts -- enrollment, in particular -- we already have KPIs in place, but these KPIs are not yet generally accessible to all decision makers. The BI committee will also help define the dashboard elements we want to create for at-a-glance purposes.

One of the great benefits of working at a college is the focus on learning, which includes the students as well as the faculty and staff. This formal business intelligence strategy is very new to all of the members of the BI team, and we're all learning as we go.

Scott Lowe is CIO of Westminster College in Fulton, Mo. Write to him at or

Dig Deeper on Small-business infrastructure and operations

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.