Correlation Analysis in My360’s Capability Analyzer is a powerful tool for understanding how different innovation capabilities relate to business success. However, correctly interpreting correlation values, selecting the right filters, and applying insights to strategy can be complex.

This expanded FAQ answers key questions, provides step-by-step instructions, and includes real-world examples to help you make data-driven innovation decisions.

Q: What is Correlation in My360, and Why Does It Matter?

A: Definition

Correlation is a statistical measure that quantifies the strength and direction of the relationship between two variables.

In My360, correlation helps organizations identify which capabilities are most strongly linked to successful innovation.

Why It Matters

Helps organizations align capabilities with strategic goals.

Identifies capability gaps that must be improved to support innovation.

Prevents wasting resources on capabilities that do not contribute to success.

Example

A manufacturing company wants to improve innovation in sustainable product development. Correlation Analysis shows:

Eco-friendly design has a high correlation (r = 0.42) with successful sustainability innovations.
Lean manufacturing has a low correlation (r = 0.15), meaning it’s not a primary driver of innovation success.

Takeaway: The company should focus on Eco-friendly design rather than trying to improve lean manufacturing for sustainability innovation.

Q: What Data is Used for Correlation Analysis in My360?

A: Important: Correlation is NOT calculated from the Primary Filter! It is calculated from the Benchmark Filter (also called the Secondary Filter).

Primary vs. Benchmark (Secondary) Filter – What’s the Difference?

Filter Type	Purpose
Primary Filter	Defines the organization(s) being analyzed. It provides the mean, min, and max scores of innovation capabilities for the selected company.
Benchmark Filter (Secondary Filter)	Defines the dataset used for correlation calculations. It determines how capabilities correlate across a broader dataset (e.g., industry, region, or top innovators).

Key Insight:

If you only analyze a single company (primary filter), you cannot find meaningful correlations.
The benchmark filter must contain a large dataset (500+ innovation assessments) for reliable correlation values.

Q: Why Should You Exclude Non-Innovators from the Benchmark?

A: Including non-innovators in correlation analysis weakens statistical accuracy.

Metaphor: Studying Bicycling Behavior

Imagine you are researching what factors help people become faster cyclists.

You collect data from experienced cyclists and analyze variables like training frequency, diet, and bike type.

Now imagine including people who do not ride bicycles.

Their data would distort the results because their speed is always 0.
This makes it harder to see patterns among actual cyclists.

In the same way, including non-innovators makes it harder to identify which capabilities drive innovation success.

How to Exclude Non-Innovators in My360

In the benchmark filter, tick the box "Exclude non-innovators" to remove companies that score too low in innovation capability.

Q: How is Correlation Calculated? (Pearson’s Formula)

A: In My360, correlation is calculated using Pearson’s correlation coefficient (r):

r=∑(Xi−Xˉ)(Yi−Yˉ)∑(Xi−Xˉ)2∑(Yi−Yˉ)2r=∑(Xi−Xˉ)2∑(Yi−Yˉ)2∑(Xi−Xˉ)(Yi−Yˉ)

Where:

XiXi = Individual values of capability scores.
XˉXˉ = Mean of capability scores in the benchmark filter.
YiYi = Individual values of strategy, leadership style, or other analyzed factors.
YˉYˉ = Mean of strategy scores in the benchmark filter.

Why Does Correlation Require a Large Dataset?

Small sample sizes lead to unstable correlation values.
A benchmark of at least 500+ assessments ensures that correlation results are statistically reliable.

If the dataset is too small, correlation may detect random patterns instead of real relationships.

Q: What Do Correlation Values Mean?

Correlation (r) Value	Interpretation
+1.0	Perfect positive correlation – The two factors always increase together.
+0.5 to +0.9	Strong positive correlation – High alignment between the variables.
+0.2 to +0.4	Moderate correlation – There is some relationship but with other influencing factors.
0	No correlation – The two variables are not related.
-0.2 to -0.4	Moderate negative correlation – When one increases, the other decreases slightly.
-0.5 to -0.9	Strong negative correlation – When one increases, the other usually decreases.

Q: Why is 0.4 Considered High in the Social Sciences?

Business performance is influenced by many factors, so correlations are rarely above 0.7.
0.4+ is considered strong in social science research because it signals a meaningful but not absolute relationship.

Q: Step-by-Step: How to Perform Correlation Analysis in My360

Step 1: Select the Area of Analysis

Go to My360’s Capability Analyzer app.
Choose Correlation Analysis from the dropdown menu.
Select an area to analyze, such as:
- Strategic Question (Why does the company innovate?)
- Type of Innovation Strategy (Incremental vs. Radical)
- Leadership Style (e.g., Spiral Staircase vs. Explorer)

Step 2: Apply the Primary and Benchmark Filters

Primary Filter: Select the organization(s) you want to analyze.
Benchmark Filter: Select the dataset used for correlation calculations.
- Exclude non-innovators for more accurate results.
- Ensure at least 500+ assessments in the dataset.

Step 3: Interpret the Correlation Table

High correlation + low internal score = Capability Gap
High internal score + low correlation = Possible misalignment with strategy

Best Practice: Prioritize improving capabilities that have high correlation but low internal scores.

Example: Correlation Analysis for Innovation Strategy

Scenario:

A company wants to determine whether it should focus on Incremental Innovation or Radical Innovation.

Results from Correlation Analysis:

Capabilities highly correlated with Radical Innovation:
Open Innovation (r = 0.45)
Risk-Taking Culture (r = 0.41)
Capabilities highly correlated with Incremental Innovation:
Process Optimization (r = 0.44)
Cost Efficiency (r = 0.38)

Interpretation:

The company has strong Process Optimization scores but low Risk-Taking scores.
This suggests Incremental Innovation is a better fit for their current strengths.
If they want to pursue Radical Innovation, they must build Risk-Taking and Open Innovation capabilities.

Decision: Adjust the innovation strategy OR develop new capabilities to support a different strategy.