The digital marketing world is abuzz with keywords like modeling, algorithms, data science, predictive analytics and data science. These keywords represent exciting new frontiers that are worthy of our time and investment but this post has been written to answer the simple question “do I need it?”.


The short answer is what you might expect “it depends…”

If You Have No Background in Statistics

If your website data exists, then you have a responsibility to look at it, and apply simple arithmetic to see what works and what doesn’t. If you are a small business and you wear several hats - then you should look at your web and marketing analytics with the keen eye of a business person.


There is a tremendous amount of insight to be extracted through simple observation coupled with the brand intimacy that you have with your business. Many questions can be answered by a simple summation of tallies, such as the below:


  • Are we up from last year?

  • What time of day are our customers most active?

  • What are the worst pages on our website?

  • Are there any devices that are having a bad experience?


These kinds of questions are simple and direct. Observe what you have on your site and correct it accordingly.


If You Have Some Grasp On Statistics

If you understand the principles of statistics, then there is a treasure trove of data to examine in platforms like Google Analytics.


Some of the data that is available in Google Analytics that is valuable is listed below:


  • Finding the correlation coefficient between items purchased (ecommerce)

  • Finding correlation between screen sizes and bounce rate

  • Using regression or multiple regression to model the value of different marketing channels (You may not have sufficient data, or even logical results depending on the sample size)

  • Review the distribution of your user engagements by binning them into standard deviations. (Useful for finding out of distribution your population has.)

  • Detect outliers using percentile or standard deviations


The tools available for statistical analysis are expansive and robust. The mainstay for many is Excel of course, but below are some tools that have been discussed heavily in analyst circles as well as some personal favorites:


  1. SPSS - This IBM tool is discussed at length in stats blogs everywhere - definitely a tool for power users

  2. SAS - A very enterprise level stats tool

  3. Minitab - Another top-shelf package, like SPSS this is a costly option

  4. Excel - Hardly needs any introduction, but there are several addons that will supercharge your excel installation

    1. RegressIt - Free and easy to use

    2. Statistician - At $20 for a license this is a great buy to streamline stats work in excel

  5. R - A powerful, free to use programming language. This is simple to grasp for analysts with nascent programming skills. There is an abundance of third party resources that are simple to integrate with it for stats, graphing, or even grabbing data using the Google Analytics Core Reporting API

  6. Python - This is free to use much like R, this is one of the languages of choice for machine learning, especially if you look into the anaconda package. This one is definitely the most complicated to set up of the bunch and may of better use if you already know how to code in at least one other high level language.



So depending on if you have no knowledge of statistics, then don’t let it stop you from looking at the data and asking questions. A tremendous amount of gold can be mined without cracking open a spreadsheet. However if you have a burning desire to dive in and learn then you are lucky, there has never been so many free, quality resources available to help someone become a whiz analyst.


If you would like to learn more about SEO, web analytics, or user experience as it regards to your website and Truvisibility you can learn more at If you are interested in trying Truvisibility out today simply go to and press the “Get Started” button, it is free and easy.