Analyze objectively

ESTIMATED READING TIME – 3:30

Learning goals for this activity

– Understand why analyzing objectively is important to talent optimization.
– Identify how to select which data to analyze first.
– Know some examples of data analytics tools that will help you uncover trends and patterns.
– Describe the four considerations that help you prioritize which problem to tackle first.
– Given some sample evidence, be able to prioritize problems in light of the four considerations.

Why analyzing objectively is important to talent optimization

Returning to our medical example, analyzing the evidence is akin to a doctor determining how to solve the problem of high cholesterol by examining blood testing results. In business terms, this is where you’ll look at the people data you’ve collected and generate a hypothesis based on your expertise.

As always, you must analyze your data within your business context. For example, you may conduct a leadership gap analysis that evaluates your leadership team’s execution style and abilities relative to your business strategy. Or you may need to work backward from a poor business result such as a slip in production quality. Remember that in business, nearly every problem is a people problem. Analyzing people data objectively uncovers issues that aren’t obvious, which allows you to quickly and effectively take action.

What data should you analyze first? Use a decision tree.

Your analysis might result in a pile of corrective actions you need to take. Don’t try to solve everything all at once. You need to prioritize. When deciding which problem to solve, take the following steps:

  • Examine the magnitude.
  • Determine the relevance.
  • Consider the breadth.
  • Look for repetition.

1. Examine the magnitude.

Some problems are bigger than others, and examining the magnitude of each allows you to separate the major problems from the minor ones. Let’s say your engagement score comes back and a mere 10 percent of your customer service reps are “highly engaged.” This is a very low score. You’ll want to act fast before turnover becomes a major issue.

2. Determine the relevance.

Relevance is where you’ll ask yourself “Does this problem really matter?” Is a suspected people problem really affecting business outcomes or employee welfare? If not, tackle others first and circle back to this one later.

Another aspect of relevance deals with not the number of employees affected but the caliber of those affected. If your high-performing employees alert you to a problem in an engagement survey, prioritize that issue ahead of another that was flagged by your lower performers. High performers who are disengaged are flight risks; work quickly to retain them.

3. Consider the breadth.

Breadth refers to how widespread a problem is. An average employee engagement score may not be concerning if it reflects a small percentage of the overall company, but if 90 percent of employees are less engaged than you’d like, this may be a systemic problem. This is something you must prioritize.

4. Look for repetition.

Look at your data and try to find repetition. Look for patterns or a theme. Is this a problem that happens again and again? Are low engagement scores more common for newer employees? Does performance slide when a manager’s team grows to have too many direct reports? If so, you’ll want to figure out the “why” and address it in short order.

What Analytic tools

You may quickly find yourself overwhelmed with the volume of people data in your organization. Organizations just starting to analyze people data might choose to use a simple spreadsheet program that includes analysis tools such as Microsoft Excel or Google Sheets in order to aggregate data and find patterns.

Smart organizations typically use more sophisticated analytics programs such as Domo or Tableau or a purpose-built talent optimization platform, which uses advanced algorithms to automate much of your analysis.

No matter what tool you choose, thoughtful analysis will help you surface underlying trends and patterns to help you make better, more data-driven decisions.