About the Episode:
In this episode of Statistically Speaking, Kerry Cunningham and Sara Boostani introduce one of the most fundamental concepts in statistics: variability. From quarterly business results to ad performance to buying cycle length, variability is everywhere — and understanding it is key to making sense of the data we encounter in B2B marketing and beyond. The episode covers what variability is, how it’s measured, and how the 6sense research team has applied it in their own studies.
Topics Covered:
- What variability is and why it matters
- Variability refers to the differences and fluctuations in whatever is being measured — and it shows up in every aspect of business and everyday life
- Key questions variability helps us answer: Why does one quarter look different from another? Why do some buying cycles take longer? Why does one ad outperform another?
- How to recognize variability in the real world
- Using weather as an approachable example: Chicago (high variability across seasons and within a single day) vs. Singapore (very low variability year-round)
- How the amount of variability in a dataset affects how confidently we can make predictions
- How statistics helps us measure variability
- There are specific methods and metrics that put a number to how much variability exists in a dataset
- Once measured, statistics can help distinguish between random noise and real, meaningful drivers of that variation.
- Real research examples from 6sense
- Marketer compensation study (2023): Variables like seniority, education, gender, and industry explained 47% of why marketer pay varies — a strong result in social science research
- Buyer experience study: A model using multiple variables explained just over 50% of why one buying cycle is longer than another, with the number of vendors evaluated and buying group size emerging as the most influential factors
Key Takeaways
- Variability is at the core of statistics — it’s simply how and why things differ from one another.
- The goal of statistical analysis is to measure variability, identify what’s driving it, and separate real signals from random noise.
- No real-world research study can fully explain 100% of variability — only perfectly controlled lab conditions might approach that, and market research certainly never would. Some factors are simply too complex or impossible to measure.
- Explaining 47–50% of variability in human behavior and business outcomes is a strong result — in social science research, explaining 10–15% is typically considered meaningful.
Related Resources
