Driving Business Outcomes with People Analytics

Most People Analytics teams are busy. Reports get published on schedule. Dashboards get built and maintained. Metrics get tracked, month after month. But there's a question worth sitting with: is any of that activity actually changing what happens in the business?

That's the challenge Steve Hall, a former people analytics leader now advising HR teams, poses in his talk on the Business Impact of People Analytics. His answer isn't "get better tools" or "add more dashboards." It's a fundamentally different way of thinking about the work. One built around business questions instead of reporting requests.

Watch Steve’s Presentation Here

The Jordan Problem: When Great Reporting Still Falls Short

Steve opens with a scenario a lot of analytics leaders will recognize.

Example: Jordan runs a well-resourced people analytics team for a large retail organization with good talent, good tools, reports delivered on time (every time). Then the CHRO calls with a real problem: store-level turnover has been climbing for eighteen months, and leadership wants it solved.

Jordan's team does what they do best. They pull the data, run the analysis, and present clean, accurate findings on where turnover is highest and who's leaving. The CHRO looks at it and says: this is great, but it doesn't answer my question. Why are they leaving? What do we do about it?

That gap? It’s between accurate reporting and an answer someone can act upon. This is exactly where most people analytics impact gets lost.

How Can People Analytics Move From Reporting to Driving Business Outcomes?

Steve's framework starts with a simple reframe: data and reporting are inputs, not outcomes. The path from data to real People Analytics business outcomes runs through two more steps that a lot of teams skip past: insights and action. The path should look like this:

1. Data & reporting: the raw material most teams are already comfortable producing
2. Insights: patterns and root causes uncovered through deliberate analysis ("this tends to lead to that")
3. Action: a specific, feasible recommendation the business can actually implement
4. Impact: the outcome the business cares about, measured on the business's terms, not the analytics team's (make sure you have a way to measure outcomes).

The trap, Hall argues, is assuming that insight and action happen automatically once a dashboard exists. Handing managers a dashboard and hoping they draw the right conclusions and act on them is a bet, not a strategy.

Insights are produced, not displayed.

Insights require reasoning which means they have someone to interpret the pattern, rule out alternative explanations, and connect it to something the business can do.

Start With the Business Goal, Not the Data You Have

Before chasing insights, Steve recommends figuring out what the business actually cares about (which isn't always obvious from inside HR).

He suggests three concrete ways to find out:

1. Talk to people outside HR. Build relationships in other departments and just ask what they're struggling with.
2. Read your company's annual report. Public companies especially lay out their priorities and challenges in writing.
3. Look at what leadership gets rewarded for financially. Compensation structures are a strong signal of what the company actually prioritizes.

Once you know the goal (for example: stock price, safety, brand reputation, engagement, retention), turn it into a testable question.

In the Jordan example, this means asking: Which locations have the best customer satisfaction scores, and why? What separates people who stay from people who leave? Why do accidents occur?

A testable question is what lets your data produce something more than a status update.

How Do You Turn People Analytics Insights Into Business Action?

Getting to actionable people analytics requires a deliberate chain, not a hopeful leap from data to decision. Steve lays it out as:

Business question → Evaluation plan → Analysis → Reasoning → Insight → Action

The evaluation plan stage matters more than it sounds like it should.

Before running any analysis ask: What data do I actually need? Should certain groups be included or excluded? Are my measures ones the business will trust and agree with?

Then, before finalizing any conclusion, Hall pushes one critical gut-check: is there another plausible explanation for what I'm seeing? Ruling out alternative explanations is what separates a defensible insight from a coincidence dressed up as a finding.

Get this wrong publicly and an analytics team burns credibility fast.

It's also worth being honest that not every good insight is actionable, and that's fine. Sometimes the data tells you something true and interesting that the business simply can't act on right now. The goal isn't a 100% action rate. It's making sure the insights that are actionable actually reach someone who can move on them. That's where making workforce insights actionable and turning insights into a compelling business story become the natural next steps once you have something real to say.

Case Study: From 250% Turnover to 65%

Steve shares a real example from a call center client with a staggering 250% annual turnover rate. This is high even by call-center standards, and expensive on every axis including recruiting costs, service quality, and eventually competitive position.

Here's how the team approached it:

• Business question: Not "predict who will quit," but "why are people leaving and what can we actually do about it?"
• Evaluation plan: They inventoried available data and tools before deciding on an approach.
• Analysis: They built a predictive model, and this is the major difference, not to forecast turnover, but to classify stayers versus leavers and surface the drivers behind the split.
• Insight: Two factors stood out: team size (a "sweet spot" around a manager-to-employee ratio) and pay, benchmarked against market data.
• Action: The business restructured teams to roughly seven employees per manager and raised pay by approximately $2 more per hour (less than the model's "optimal" number, but what the business could actually afford).

The result: turnover dropped from 250% to 65%. Nobody needed a perfect answer. They needed one that was both grounded in evidence and feasible to implement — which is the entire point of the framework.

Why Teams Get Stuck as Order-Takers

If this framework is so clearly valuable, why don't more teams operate this way? Steve points to a few recurring barriers:

• Tactical overload. Ad hoc requests and recurring reports eat the time that would otherwise go toward the evaluation-plan-and-analysis chain.
• Isolation from the business. Without regular conversations outside HR, it's hard to know which questions actually matter.
• Resource constraints. Limited technical capability or data access can cap how far a team can push into causal analysis.

His prescription is a shift in posture across four areas:

• Democratize access to data. Self-service reporting frees the team from being a report factory and creates room for higher-value work.
• Build research design and analytics capability. Teams need people who can design a rigorous evaluation plan, not just build a chart.
• Adopt an insight-driven mindset. When someone asks "has headcount changed?", ask "why do you want to know?" That's the question behind the question — and it's usually the thing worth actually answering.
• Focus on what the business cares about. Answering the wrong question well still produces no value.

The Bottom Line

The core message is refreshingly practical: impact doesn't fall out of a dashboard. It comes from intentionally connecting a business goal to a testable question, running a disciplined evaluation, and landing on a recommendation the business can realistically act on.

As Hall puts it: don't wait for your CHRO to say "I already knew that." Make sure they never have to.

Ready to see what One Model can do for your business?

Connect with us today!