Asking Better Questions in a Data-Rich World

Asking Better Questions in a Data-Rich World

By the time most executives open their dashboards or data lakes, they’ve already formed a hypothesis — and often a bias. Yet, paradoxically, as organizations collect more data than ever before, decision quality hasn’t improved proportionately. The yawning gap between data abundance and decision impact often comes down not to analytic horsepower but to a far more human skill: asking the right questions.

The Data Paradox: More Isn’t Always Better

The global big data market is projected to exceed $100 billion by 2027, with double-digit growth as enterprises increase investments in analytics platforms and capabilities. Yet McKinsey’s research shows that while over two-thirds of executives say decision processes in their organizations are becoming more data-driven, fewer than half report that decisions are timely — and over 60 % say at least half of their decision time is ineffective.

This signals a core issue: volume of data is not a proxy for clarity of thought. In many cases, leaders fall into the trap of data-driven inertia — spending time mining for insights before they are clear on the decision to be made.

Start With the Decision, Not the Data

Researchers at Wharton refer to this disconnect as the distinction between data-driven and decision-driven analytics. The latter flips the traditional analytics process by starting with a decision and then asking what data is necessary to inform that decision.

Consider the case of HP’s Instant Ink business. Faced with rising subscription cancellations, a naive data-driven approach might ask: “Which customers are most likely to churn?” Instead, HP leadership reframed the question to evaluate the impact of incentives: “Which customers are most likely to stay if offered a targeted incentive?” Answering this required randomized experimentation to assess causal effects, not just predictive correlations — a subtle but vital shift in question framing that directly informed strategy.

The Sophistication Gradient of Questions

Researchers distinguish between types of questions executives should ask in a data-rich world:

• Factual questions: What happened, and what is likely to happen next?
• Counterfactual questions: What would happen under alternative scenarios?
• Causal questions: What factors influence outcomes and why?

In high-stakes contexts, counterfactual and causal questions often yield far richer strategic insights than surface-level correlations.

From Theory to Practice: Case Examples

1. People Analytics at Google

One of the earliest and most cited successes in modern analytics is Google’s Project Oxygen. By analyzing over 10,000 performance reviews, the company identified behavioral patterns that distinguished high-performing managers. The findings informed training programs that lifted median managerial favorability scores significantly.

The critical success factor wasn’t just analyzing data but asking which behaviors matter most and aligning the question with organizational goals — effectiveness, retention, and performance.

2. Starbucks’ Location Analytics Strategy

After a wave of closures in 2008, Starbucks began to employ advanced analytics to identify where new stores would succeed. Rather than simply mapping high foot traffic, the company asked: “What combination of demographic, traffic, and competitive variables best predicts success?” The resulting model incorporated diverse datasets — including regional insights — which enabled more strategic expansion decisions.

3. Political Campaigns and Targeted Persuasion

Perhaps one of the most illustrative counterfactual studies came from the 2012 U.S. presidential campaign. Instead of asking who was most likely to vote for a candidate, the Obama campaign asked: “Who is most likely to change their vote if engaged?” This counterfactual framing drove highly efficient targeting that influenced the outcome.

The Psychology of Questioning Under Data Overload

Behavioral scientists emphasize that better questions begin with curiosity, not certainty. Harvard researchers have shown that experts who intentionally ask follow-up or probing questions not only gather deeper insights but also build trust and collaboration.

This aligns with Harvard Business Review’s broader research on strategic questioning: more effective leaders don’t simply provide answers — they stimulate deeper thinking in others through better inquiry.

Why Simple Questions Are Often Hardest to Ask

In data science practice, many early failures occur because the initial question is too broad or ill-defined. For example, asking “Which vaccine is better?” without specifying the performance metric (immunity, transmission, side effects) yields ambiguous and unhelpful insights. Businesses experience similar friction when analysts are given vague problems, such as “Improve sales,” without context or priorities.

The solution is to craft answerable questions — those that are specific, measurable, and tied to a decision framework.

Structured Frameworks for Better Inquiry

Adopting a hypothesis-driven or structured analytics process like BADIR (Business question → Analytics → Insights → Recommendations) helps ensure that analytical efforts remain aligned with strategy and outcomes, not just technical outputs.

Another practical principle is the “Expected Value of Perfect Information” concept from decision science — a good question should be one where the expected answer will materially affect decision quality.

Barriers and Behavioral Challenges

Despite the promise of better questions, organizational culture often resists change. A McKinsey survey found barriers like an entrenched preference for experience over data and analytical skills gaps among decisionmakers.

Leaders must address these human and structural challenges by incentivizing learning, adjusting roles for data literacy, and teaching teams how to ask and refine critical questions.

Conclusion: From Data Rich to Insight Driven

In a world where data generation outpaces human capacity to interpret it, the most valuable skill is discernment — the ability to translate information into meaningful decisions. Asking better questions isn’t optional; it’s foundational to unlocking the value buried within data.

Whether guiding a boardroom strategy or optimizing daily operations, reframing problems, challenging assumptions, and clarifying what we need to know — not just what we can measure — will separate leaders who thrive from those who merely survive in the age of data.

Follow us on social media for more updates: Facebook | X | Instagram | LinkedIn | YouTube | Pinterest | Mastodon | Bluesky