Do Digital Marketing Like a Statistician

What’s your opinion about statistics? Too often I think stats gets a bad reputation, either being regarded as too complicated or not useful enough. I need to tell you this: you can understand the basic principles of stats, and they will make a huge impact on your digital marketing efforts.


Why is Statistics Important to Digital Marketers?

Digital marketers need statistics to be able to make confident decisions regarding their work. Statistics will help marketers to analyze the data and make accurate recommendations to clients and management.

I have seen and used applications for these digital marketing practices:

  • Outreach & Email Marketing: You can use what you learn in the blog post to test different versions of an outreach email to see if you should phrase your emails differently or not.
  • PPC and Social Media Ads: Digital marketers use statistics principles to analyze the data from their ad campaigns and improve their accuracy of customer segments.
  • Google Analytics: Google Analytics will be used more effectively if you understand different ways you can analyze your website traffic.
  • Conversion Rate Optimization (CRO): Statistics will help digital marketers run more effective A/B tests. A/B testing is a common practice within the discipline of conversion rate optimization.
  • Social Media Engagement: Perform some A/B tests to test different templates of a social media post. For example, you’ll determine whether asking questions is more engaging or not.

What is Statistics?

Statistics is learning about the world around us to make informed decisions. In digital marketing, the world we want to learn about is a company’s current and potential customers.

In statistics, we call this group a population. Usually, studying an entire population is very difficult. Therefore, a marketer will take a sample that is large enough for them to make informed statements about the population. Then, they will perform a study on the sample.

Using Statistics in Digital  Marketing

The effects of sample size and population size (Outreach, Social Media Ads, PPC, Email Marketing, CRO)

Each dataset can be summarized with a mean (an average) and a standard deviation. If you aren’t familiar with how to find the standard deviation of a dataset, this resource is very helpful.  In statistics, we use the mean, standard deviation and the sample size to learn about the population.

For example, if you were trying to learn about the demographics of a website’s visitors, you would take a sample from the group and study the data or perform an experiment. Usually, datasets with small sample sizes are less representative of the population.

In our example, if you collect a sample of 100 website visitors, learning about all of the customers will be a lot harder compared to using a sample of 5,000 visitors. Samples with a lot of data, i.e. 5,000 people, have a mean and a standard deviation that’s closer to the population’s. Therefore, you can make informed statements about a population if you have enough data. See the figures below.


In the digital marketing world, you need to be aware of the amount of data you’re working with. If your social media ads aren’t getting enough impressions, you will need to be patient until you get enough data. If you reach out to sites for guest posts or run an email marketing campaign with a certain strategy, you need to test the strategy long enough to get a necessary amount of data (like 80 to 100 emails). In other words, just because a new strategy is highly successful after 10 trials doesn’t mean you should adopt the new strategy.


A/B Tests and Other Experiments (CRO, Google Analytics, PPC ads)

Here’s another example. CRO is a common practice within digital marketing. The purpose of an A/B test is to find what factor makes a significant impact on online conversions. One example is “does a new call to action make a significant difference?”. In order to perform the experiment, you need to define a hypothesis, run the experiment, collect data, get the sample’s mean and standard deviation, and then receive a p-value.

  • P-Value: The probability of getting results as extreme or more extreme than a value if the original hypothesis we make is true. You get the p-value from the new data in your study. In the tools found at the bottom of the article, the p-value is given to you after you input the necessary information.


Then, you see if the p value is less than a level of significance you determine before you perform the study. If the p value is less than the level of significance, the new data is significantly different than before.

  • For your information, the level of significance is one minus your desired confidence level. The most common confidence level is 95%.

Therefore, you found what factor improves your online conversions.



What do you do if your P-value isn’t significant? You can still learn from your results. Open this resource for more details.

Hopefully, I’ve given you enough to work with. If you have more questions about statistics and how they relate to digital marketing, check out these other useful resources which, are helpful in applying the discipline of statistics to digital marketing.

Other Resources

Kevin Toney
SEO Specialist
Kevin Toney is an aspiring business mind that is passionate about accomplishing the goals he has set for himself. He loves data, and is very interested to see how it influences decision making. When he is not working, he loves to make friends and to be outdoors.