Welcome to "Think Like an Analytics-Savvy Manager." Every day, managers face requests that sound reasonable but lead nowhere productive. Someone asks you to "look at the data and tell me what's going on" or to "pull some numbers before the meeting." These vague asks consume time, frustrate, and rarely drive real decisions.

Throughout this course, you will learn to think like an analytics-savvy decision-maker. That doesn't mean becoming a data scientist—it means knowing how to frame the right questions, partner effectively with data experts, follow a simple process that keeps analysis focused, and balance your experience with evidence.

You'll use concepts from the HBR Guide to Data Analytics Basics for Managers to learn how to transform vague requests into sharp decision questions. From there, you'll practice partnering with analysts and data scientists so projects stay aligned and efficient. You'll also follow a repeatable analytics process that helps teams avoid common pitfalls. Finally, you'll learn to blend intuition with evidence, knowing when to trust your gut and when to wait for more data.

The single biggest waste in business analytics is analysis that doesn't connect to a decision question. When someone says "Analyze our customer engagement" or "Look at last quarter's performance," there's no clear action at the end. You might spend hours pulling data, only to hear "Interesting, but what should we do with this?"

Before any analysis begins, identify the decision that needs to be made. A decision question has a specific choice attached to it. Consider the difference between a vague request like "Can you look at our campaign data?" and a decision-focused alternative such as "Should we shift budget from email campaigns to paid social for Q3, or keep the current allocation?" The second version tells you exactly what choice is on the table, narrows the scope of what data matters, and makes it clear what "useful" analysis looks like.

To reframe a vague request, ask yourself what someone will actually do differently based on this analysis. If you can't answer that question, you don't have a decision question yet. Your job is to anchor every analysis to a specific choice, ensuring that the work you do directly informs a real business decision.

Once you have a decision question, you need three more pieces before diving into data: success criteria, stakeholders, and constraints. Each of these elements shapes the analysis and prevents wasted effort down the line.

Success criteria answer the fundamental question of what "good" looks like. If you're deciding whether to expand a pilot program, you need to know what results would justify that expansion. Perhaps it's a 10% lift in conversions, or maybe it's achieving positive ROI within six months. Without defining success upfront, you'll end up debating interpretations after the analysis is done—and that's a recipe for indecision. Establishing these criteria early creates a shared understanding of what you're trying to achieve.

Stakeholders are the people who care about the outcome and have influence over the decision. Knowing your stakeholders helps you tailor the analysis and anticipate objections. For instance, if the CFO cares about cost efficiency while the VP of Sales focuses on pipeline volume, your analysis should address both perspectives. Ignoring a key stakeholder often means redoing work later when their concerns surface.

Constraints represent the practical limits on your analysis, including timeline, available data, budget, or regulatory restrictions. If someone needs an answer by Friday and your ideal dataset takes three weeks to compile, you need to know that before you start.

Consider a scenario where someone asks "Should we expand our customer loyalty program?" Before analyzing, you'd clarify what metrics define success—retention rate, revenue per customer, or program cost. You'd identify who needs to approve the decision, whether that's marketing, finance, or operations. And you'd determine when the decision is needed and what data you actually have access to. Answering these questions upfront prevents wasted cycles and misaligned expectations.

Even with the best intentions, requests often arrive incomplete. This is where it is helpful to formulate a short set of clarifying questions that you ask before agreeing to any analysis. These questions protect your time, sharpen the project, and build trust with stakeholders who will appreciate your rigor.

Start by establishing the decision itself. Ask "What decision are we trying to make?" and "What are the options on the table?" These questions force the requester to articulate the choice, not just the curiosity. Following that, surface the success criteria by asking "What would success look like—what result would make us confident in one option over another?"

From there, move to timing and scope. Questions like "When do you need this, and what's driving that deadline?" and "Are there specific segments, time periods, or data sources we should focus on—or exclude?" reveal constraints and help you avoid the trap of analyzing everything when only a slice matters. Finally, check for hidden stakeholders by asking "Who else needs to weigh in on this decision, and what do they care about?" You'd be surprised how often a forgotten stakeholder derails a project at the last minute.

Here's how this might sound in practice when a people manager receives a vague request:

• Nova: Hey, can you pull some data on our team's engagement scores? I need it for a meeting tomorrow.
• Chris: Sure, I can help with that. What decision are you trying to make with this data?
• Nova: Well, I've been hearing some complaints, and I want to see what's going on.
• Chris: Got it. Are you deciding whether to change how we run team meetings, or maybe looking at workload distribution, or something else?
• Nova: Actually, I'm trying to figure out if we should restructure our one-on-ones. Some people say they're not helpful.
• Chris: That's much clearer. So we're deciding whether to keep, modify, or replace our current one-on-one format. What would success look like—higher satisfaction scores, more actionable feedback, something else?

Notice how Chris's clarifying questions transformed a vague request into a focused decision question. In just a few exchanges, they moved from "pull some engagement data" to a specific choice about one-on-one meetings, with success criteria starting to take shape. The conversation only takes a minute or two, but saves hours of unfocused analysis.

Asking these questions might feel like slowing things down, but it actually speeds up the overall process. A 10-minute clarifying conversation can save days of unfocused work, and it positions you as someone who delivers analysis that actually drives action. In the upcoming role-play session, you'll practice using these clarifying questions in a realistic scenario where a colleague comes to you with a vague, urgent request. You'll work through the conversation to transform that request into a focused, decision-ready analysis scope.