Creating an effective business intelligence process

Business intelligence is intended to facilitate fact-based decision-making. BI employs technology, software applications, and practices for the collection, integration, analysis, and presentation of business information to support decision-making. Tools used include traditional query, reporting, and OLAP functionally as well as data mining, dashboards, scorecards, and visualisation.

Articles in the press oftentimes speak to BI technology while ignoring the BI process itself. Technology and applications are often over emphasised and the notion that BI is a practice is lost in the shuffle. This is potentially problematic, because BI can only provide value when its contribution is understood within a known and understood context. What is presumed is that firms inherently understand and have defined what their BI process should look like such that roles and responsibilities are clearly defined and processes and desired outcomes are clearly understood by all parties involved in the decision making encounter.

Circumstances often define the nature of the decisions to be made, the role of BI, the relevant stakeholders, and the value sought in the decision-making activity. However, if the decision-making process is ill defined and unstructured, BI efforts can be misapplied, incorrectly appropriated, or even ignored. This is dangerous not only to the effectiveness of the decision-making activity, but to BI itself because it isn’t demonstrating its potential benefit to line-of-business or other functional area managers.

To address this issue, it is critical that BI analysts and decision makers sit down and clearly articulate frameworks for decision making that define goals and objectives for the process, measures to be used, player responsibilities, and potentially even actions that should be considered or taken based on BI analytical outcomes. This structured process should provide the context wherein everyone involved understands the nature of his or her contribution to the decision-making process and the relevance of BI to the process itself.

This idea may seem cumbersome or unnecessary to some, but I can tell you that this type of decision-making process framework is routinely and successfully used in another discipline – medicine. Medicine is focused on generating successful outcomes for a patient client. It employs technologies and applications for discovery, diagnostics, monitoring, and treatment. Use of BI-type tools and services are often requested by physician generalists, who hand off to knowledge specialists who use technology to perform diagnostic tests, interpret their output, and inform the decision maker as to their significance and meaning. These series of activities help to identify the specific nature of the problem or issue as well as the action(s) to be taken. What is consistent here is the contextual process in which the technology and analytics are framed.

Knowledge exchange protocols – SOAP

Knowledge exchange protocols improve sensemaking by embedding BI-type analytical processes within a formal structure that helps to legitimize the role of analytics in decision-making. The goal is for all decision-making stakeholders to buy in to a decision-making process that informs participants as to their roles and responsibilities. Diagnostic tools may then be applied within this context.

Knowledge exchange protocols have been used in the medical community for some time and most patients are at least subtly aware of their existence. SOAP is a knowledge exchange protocol used to structure and document situation-oriented, physician/patient clinical encounters. SOAP (subjective, objective, assessment, plan) provides a framework for:

• structuring clinician-patient exchanges,

• framing the clinician’s thinking about perceived problems and issues,

• documenting techniques and tests employed by the clinician in the knowledge creation/decision making process, and

• sharing the clinician’s reasons for actions taken to address patient issues.

As a result, SOAP process provides a consistent process for defining and documenting:

• what the physician understands about their patient’s situation (sense making activities),

• how the physician closes gaps in their understanding about the patient’s situation (knowledge creation), and

• what actions the physician takes relative to treatments (decision making).

In the subjective phase, the patient describes the problem to the physician (the reason for their visit). In the objective phase, the physician requests objective tests that are often performed and interpreted by knowledge specialists (e.g., x-ray, CAT Scan, CBC, etc.). Based on the patient’s complaints and the test results, the physician makes an assessment (diagnosis) leading to a plan of the action (more tests, prescriptions, nothing, etc.).

What is especially important about knowledge exchange protocols is that all parties involved buy in to them in order to effectively contextualise and legitimise the decision making process. In the table below, the uses of SOAP for medical and potentially for BI applications are illustrated. While not necessarily understanding the tests themselves and the clinician’s explicit and tacit reasoning, the patient (client) understands the relevance of the problem solving process and its importance to the decision making of the physician (analyst), whose judgment they typically must trust. The objective underlying this process (to heal the patient) and the roles of everyone involved is understood and accepted by all parties.

Comparing medical practice and business intelligence as knowledge exchange protocols

Medical students’ learning of the SOAP protocol is an important component of the clinical teaching process. Not only is SOAP used to configure the clinician-patient encounter, but it also serves to structure the documentation found in the patient’s medical record. This record can then be used to assess how clinician knowledge has deployed and manifested over time. For example, analyses of SOAPs by agencies such as the Joint Commission on the Accreditation of Hospitals (JCAH) have been used to gain an overall impression of the nature and quality of patient care at various medical institutions. Put in a business intelligence perspective, the SOAPs can be audited to gain knowledge about how decision maker perceptions are created and justified. Formal review processes use SOAP documentation to evaluate clinician assumptions underlying action, resulting in analytic processes and outcomes coming into focal awareness and scrutiny. The linkage between data, analysis, and outcomes is made transparent, allowing analysis of variables and their relationships that have been used to determine causality and decision-making. This allows a retrospective assessment of the effectiveness of the procedures and analytics employed.

The discussion of the SOAP knowledge exchange protocol example suggests not just the appropriateness of incorporating BI in knowledge-driven decision-making processes, but the value of clearly delineating a contextual process wherein the utility of BI analytics makes sense to stakeholders, even if only an expert can interpret their meaning. In effect, knowledge exchange protocols are used to provide functional legitimacy to a process wherein the analytics are imputed with both relevance and credence that facilitates evidence-based decision-making and action.

In many business schools, students are introduced to an assessment protocol called SWOT that stands for strengths, weaknesses, opportunities and threats.

• S trengths: attributes of the organisation that are helpful to achieving the objective.

• W eaknesses: attributes of the organisation that are harmful to achieving the objective.

• O pportunities: external conditions that are helpful to achieving the objective.

• T hreats: external conditions that could damage performance.

SWOT is a method used to structure the analysis of a firm’s situation relative to some issue, project, or business venture. SWOT provides an effective orientation for communicating internal and external factors that are favourable and unfavourable to achieving an objective. However, if the SWOT does not start by defining a desired end state or objective, it runs the risk of being useless.

While SWOT is another example of a knowledge exchange protocol, I assert that the SOAP protocol is better suited to BI efforts because it is specifically decision focused and it follows a more intuitive problem solving structure. Moreover, the configuration of this protocol forces the objective to be captured in the Subjective phase of the analysis and all subsequent activities in the protocol address that objective.

As in medicine, use of a SOAP-like protocol in BI activities can help firms to document their contribution to decision activities over time. Firms can use them to gain knowledge about the effectiveness of BI-based decisions and to assess opportunities to deploy new BI-related tools and techniques. In medicine, a formal medical review process uses SOAP documentation to evaluate clinician assumptions underlying action, resulting in tacit knowledge coming into focal awareness and scrutiny. This activity, called “Grand Rounds” in medical circles, provides a vehicle for auditing decision processes and outcomes. In surgical grand rounds, for example, new clinical and scientific information is presented and new procedures for patient care are explained. In addition, existing or historical cases can be scrutinised to see what could be done differently to potentially enhance outcomes. With a focus upon improving decision-making activities, BI-related efforts could adopt a similar ‘rounds’ technique to foster ongoing scrutiny of decision-making performance.

By using a protocol like SOAP, the linkage between issues, data, analysis and outcomes is made logical and transparent. It can help to document how BI contributes to decision-making and its impact on outcomes.

User acceptance of SOAP-like protocols is typically high if introduced correctly. Many clients are uncertain what BI is or how it fits in with there decision making activities. Conversely, many BI experts are uncertain about how and when to explain BI results to decision makers or even when to offer their services at all. Implementing a knowledge exchange protocol such as SOAP helps both parties by providing a common orientation and structure that defines and assigns roles and responsibilities in a problem solving process.

No one today would entertain seeing a physician who used anecdotal evidence and ad hoc heuristics to solve their medical problems. Nor should a business employ these same techniques, especially when they can avail themselves of so much data and analytical capabilities. BI’s value is that it can contribute to evidence-based decision-making. However, its utility can only be appreciated and it’s findings accepted when all decision stakeholders understand and appreciate the nature of its role relative to the decision-making process as well as to the intended objective(s).

What I argue here is that if you can provide your people with this type of structured context, then they should be more willing to accept BI’s potential utility in decision making activities. Use of knowledge exchange protocols can help BI efforts be appreciated as the enabling set of technology-based diagnostic tools and processes it should be, rather than the complex and mystifying set of geeky technologies that many perceive it to be.

Dr. Richard Herschel is Chair of the Decision & System Sciences Department in the Haub School of Business at Saint Joseph’s University in Philadelphia. Dr. Herschel worked as a systems analyst at Johnson & Johnson, Columbia Pictures, Schering-Plough Corporation, and Maryland National Bank. Dr. Herschel writes extensively about knowledge management and business. He is the Education Channel expert for the Business Intelligence Network (http://www.b-eye-network.com/)

Richard T. Herschel, PhD

Saint Joseph’s University

Philadelphia, PA USA

herschel@sju.edu