Data Intelligence Monthly
Issue 03 - January 2026
Welcome to the third issue of Data Intelligence Monthly. Each month, we will touch-on a specific topic along the data analytics life-cycle. In this issue, we discuss an important primary step in data analytics: Identifying Research Questions (RQ).
Research Questions (RQ) as a Core Technique in Data Analytics
Turning Business Curiosity into Analytical Insight
In modern data analytics, sophisticated tools and vast datasets often receive most of the attention. Yet, one of the most powerful—and frequently underutilized—techniques is far more fundamental: identifying research questions (RQ).
Well-designed research questions act as the connective tissue between business strategy and analytical execution. They ensure that analytics efforts remain focused, relevant, and decision-oriented.
Defining the Problem or Opportunity
The first and most important role of a research question is to clearly define the problem or opportunity. Organizations often approach analytics with vague concerns: declining margins, rising risk exposure, or slower asset growth. Without a precise problem definition, data analytics risk may become an exercise solely in data exploration rather than decision support.
A strong research question reframes symptoms into a testable problem(s).
Example: A retail bank notices an increase in credit card delinquencies. Rather than asking, “Why are delinquencies rising?”, a more analytically and focussed research question might be: “To what extent are recent increases in delinquency rates driven by changes in credit profile versus changes in overall macroeconomic conditions?”.
Takeaway: This reframing narrows the analytical scope, identifies causal pathways (diagnostics), and establishes clear expectations for the type of data and analysis required.
Understanding Stakeholders
Research questions also serve as a translation layer between technical analysts and business stakeholders. Different stakeholders—executives, finance and risk managers, compliance officers, portfolio managers—often view the same problem through different lenses. Effective research questions reflect these perspectives explicitly.
By clarifying who the analysis is for and what decision they need to make, research questions help prioritize relevance over technical elegance.
Example: An asset management firm considering a shift toward ESG-weighted portfolios may involve stakeholders with competing priorities. Portfolio managers focus on maximizing returns while managing risks, compliance teams on disclosure obligations, while investment managers may centre on the broader view (sustainability alignment and return). A stakeholder-informed RQ may be: “How does integrating ESG scores into portfolio construction affect risk-adjusted returns across client segments?”
Takeaway: This question acknowledges multiple stakeholders needs and ensures the analysis produces insights that can actually be acted upon.
Assessing the Current State
Before projecting change or recommending action, analytics must establish a clear understanding of the current state. Research questions guide analysts toward the right baseline metrics, timeframes, and benchmarks.
Without a well-defined current state, future comparisons and trend analyses lack credibility.
Example: A commercial bank wants to improve loan approval turnaround times. Instead of immediately modeling process improvements, a foundational research question might be: “What is the current distribution of loan approval times by product type, region, and customer risk tier?”
Takeaway: This question drives descriptive analytics that reveal bottlenecks, variability, and performance gaps—critical inputs before any predictive or prescriptive analysis can occur.
Projecting the Future State
One of the most valuable contributions of analytics in finance and investment is its ability to anticipate future outcomes. Research questions help structure forecasts, scenarios, and stress tests by clearly defining what future state matters.
Well-crafted research questions ensure projections remain grounded in business relevance rather than abstract modeling.
Example: A pension fund evaluating its long-term asset allocation may ask: “Under different interest rate and inflation scenarios, how is the funded status of the plan expected to evolve over the next ten years?”
Takeaway: This research question explicitly connects future uncertainty to a measurable business outcome, guiding the selection of models, assumptions, and scenarios.
Bringing It All Together
Only after defining the problem, understanding stakeholders, assessing the current state, and considering the future state should analysts formulate the final set of research questions. At this stage, questions are refined to be specific, measurable, and analytically feasible. In practice, this often results in a hierarchy of questions—primary questions supported by secondary (diagnostic) questions.
Research questions are not merely a preliminary step in data analytics—they are a discipline that shapes the entire analytical lifecycle. Research questions ensure analytics efforts remain aligned with real business decisions.
Ultimately, strong research questions transform data from a passive asset into an active decision-making tool—ensuring that analytics answers not just what the data says, but what the business needs to know.
If you are interested in discussing, planning or developing your data analytics strategy, please contact us for a free 30-minute consultation.