Get to Know the Shape of your Marketing Data Better
Find out how your marketing data ‘looks’ through Variance and Standard Deviation.
Have you ever wondered how your data looks? Probably not. But after today, you’ll be asking yourself how spread out your data is — and you’ll start thinking of its “shape” like it’s a real person. (Cue “Shape of You” by Ed Sheeran 🎶)
Summary Statistics
Marketing data can range from just 20 entries to hundreds of thousands, making it tough to analyse. That’s why one of the first things you do when exploring data is using summary statistics — simple tools that give you a snapshot of your data without getting lost in the details.
These are some of the core measures you’ll use to understand your data:
- Measures of central tendency: Mean, median, mode (basically, the “middle” of your data).
- Min and Max values: The highest and lowest values in your dataset.
- Count: The number of unique values.
These are great for getting an overall sense of where your data sits. But they don’t tell you much about how your data is spread out. That’s where measures of dispersion come in.
How Does Your Data “Look”?
When I say the “shape” of your data, It’s not about pretty visuals (though, we’ll get there!), it’s about how spread out or clustered your data points are. Are they tightly packed around the average value, or are they scattered far apart? Are there any extreme values?
This is where variance and standard deviation come in. These two metrics help you understand how spread out your data points are and give you an idea of where most of your data falls. Don’t worry, I’ll break it down into simple steps.
Copy the dataset to follow along and view the calculations I’ll be explaining below.
There’s also a quick tip inside on how you can find all this data in just 2 clicks!
Measure #1: Range
This is the simplest way to understand the data dispersion. The range is the difference between the highest and lowest values. So if your marketing data includes revenue per customer, the range would tell you how much your highest-paying customer differs from your lowest-paying customer.
For example, if your lowest revenue is $100 and your highest is $1,000, your range is $900. This tells you how much variation exists, but doesn’t give you much beyond the extreme ends.
How to calculate in Google Sheets:
=MAX(range)gives you the highest value.=MIN(range)gives you the lowest value.- Subtract the minimum from the maximum to find the range.
Measure #2: Interquartile Range (IQR)
The IQR shows you the spread of the middle 50% of your data. Think of it like a year divided into 4 quarters (quartiles) and focusing on the middle two. This helps you focus on where most of your data falls and ignore any outliers (extreme values that may be errors or not useful).
(Image of 4 quarters and bracket around Q2 & Q3)
The IQR helps with customer segmentation. You can divide your customers into:
- Lower quartile (bottom 25%)
- Median (middle 50%)
- Upper quartile (top 25%)
How to calculate in Google Sheets:
- Use the
=QUARTILE(range, 1)and=QUARTILE(range, 3)functions to find the first and third quartiles, then subtract them to get the IQR.
I hope I haven’t lost you yet! I promise this will all make sense in the end.
Measure #3: Variance of Marketing Data
Variance measures how far your data points are from the mean (the average). It’s useful for understanding whether your data is tightly grouped or widely spread out, but by itself, variance can be a bit hard to interpret.
For example, if you’re tracking your revenue per customer, variance will tell you if most of your customers are spending close to the average, or if some are spending much more (or less) than others.
How to calculate in Google Sheets:
- Use the
=VAR(range)function to find the variance of your data.
Measure #4: Standard Deviation
The standard deviation is simply the square root of the variance, and it’s much easier to understand. It tells you how much your data falls within certain ranges around the average.
It breaks your data into chunks being:
- 68% of your data falls within 1 standard deviation of the mean(average)
- 95% falls within 2 standard deviations
- 99% falls within 3 standard deviations
In marketing, this could help you understand things like:
- How much variation there is in customer spend across different segments.
- Whether your campaign results are stable or if there are big swings in performance.
So, if you’re analysing marketing metrics like customer revenue, knowing the standard deviation helps you see how much the typical customer’s revenue differs from the average customer.
Example: Let’s say you have a customer revenue average of $500, with a standard deviation of $50. That means:
- 68% of your typical customers will have a revenue between $450($500 — $50) and $550($500+$50). We are subtracting the standard deviation from the average to find the lower amount, and adding to get the higher range.
- 95% of your typical customers will have a revenue between $400($500–$100) and $600($100). Because this falls with 2 standard deviations ($50+$50), were running the calculation with $100.
- 99% will have a revenue between $350($500–$150)and $650(500+$150). Here, $50×3(standard deviations) is $150.
How to calculate in Google Sheets:
- Use the
=STDEVP(range)function to find the standard deviation of your data.
Visualising Standard Deviation with The Normal Distribution
Standard deviation gets even clearer when you visualise it with a normal distribution — or a bell curve. This curve shows that most of your data falls near the mean, and fewer data points exist the further you get from the average.
You can do the easily with a histogram on Google Sheets. Imagine a smooth curve on top of the bars, and you’ll see the a normal distribution like the illustration below.
This concept is important because it helps you understand whether your data is normally distributed or if it’s skewed. With marketing data, understanding the distribution of your metrics can help you make better predictions and more accurate decisions.
Now that you know how to use variance and standard deviation, you’re ready to start looking at your data in a whole new way!
If you’re ready to dive deeper into your data, check out my upcoming post on types of data distributions — where I’ll cover everything from normal distributions to skewed data and how to use them in your analysis.
