thesmarthomeninja / mmm Goto Github PK

Source: http://datafeedtoolbox.com/marketing-mix-model-for-all-using-r-for-mmm/

Do you want to know how much you need to spend vs how much revenue you will get back with your advertising campaigns? i.e. shopping or search campaigns at a high level for a specific period of time?

Marketing Mix Model for All: Using R for MMM June 1, 2017 Credit Goes To: Jessica Langford Machine Learning, Optimization, R, Statistics Understanding the ROI across all of your paid marketing channels is a top priority for senior-level executives across every industry and every geographical market. Getting a clear sense of the ROI on each channel allows companies to answer really important questions. For example:

What will happen if I increase my Email spend by 20%? What is the level of saturation I can expect from my Paid Search channel? How do I incorporate seasonality into my budget allocation strategy? How do I optimize my budget allocation across all of my paid channels? Which geographies should I focus my efforts? These are consequential questions that have the potential to have a major impact on how a business operates. With that, it’s no wonder that companies devote significant time, energy , and resources to creating (or purchasing) a Marketing (or Media) Mix Modeling framework. The aim of this post is to show you how you can use the data that you already have in conjunction with open source tools to create your own MMM solution.

THIS IS SUPER SIMPLE. YOU JUST DOWNLOAD YOUR CAMPAIGNS DATA AT A VERY HIGH LEVEL. YOU NEED TO HAVE YOUR CAMPAIGN NAME, CATEGORY, SPEND/COST AND REVENUE. THATS IT!

Once you have downloaded your CSV file in the format listed in the article that this script was sourced from... Then you want to basically follow the steps in the article and go ahead and grab the full script in MMMv1 file rather than copy and pasting all of it. There are customizations you can add to the parameters within the script so pay close attention to the article. Happy Scripting you R Ninja you.