Predictive Analytics, a solution to knowing your Marketing ROI

There is an old adage in marketing that goes “Half the money that I spend in advertising is wasted; the trouble is, I don’t know which half”. This has often been attributed to department store magnate and father of modern advertising John Wanamaker. William Lever of Unilever fame also said something very similar. These two gentleman were masters of marketing, yet they struggled with knowing which of their marketing campaigns worked and which didn’t.

Today, marketeers are still struggling with the same problem as Wanamaker and Lever were over 100 years ago; how much return on investment (ROI) do I get for every dollar spent on a marketing campaign? Believe it or not, calculating marketing ROI is a very difficult tasks because the complexities involved in calculating the amount of sales that is actually generated by a campaign. When attributing sales revenue as a direct effect of a campaign, you must take into account many factors including seasonal sales cycle variations, competitor behaviour and other overlapping sales campaigns. The complexities are such that traditional data analytics that are used by many agencies and organisation cannot accurately calculate marketing ROI at all. However understanding your marketing ROI is essential to knowing where best to invest for maximum returns.

Up until now marketing managers have been relying on guestimates of ROI and their experience (AKA “gut feel”) when making decisions on where to spend their money. But obviously this is very suboptimal and heavily dependent on how good that “gut feel” is! In today’s corporate environment where boards are demanding better metrics on marketing effectiveness and CFO want justification on marketing budgets, this is often not good enough.

However with Predictive Analytics and Machine Learning, there are now solutions to this problem.

Predictive Analytics is the science of using techniques such as machine learning to find patterns in historical data to build models and make future predictions. Machine learning is the study and construction of computer algorithms that can learn from and make predictions on data.

Confused? Well, one simple way to explain machine learning is to imagine you had a large spreadsheet (with at least several hundred rows of data) that has a number of columns. If you were to feed the data of the spreadsheet into a machine learning program, it would be able to “learn” from that data and internally build a computer model. If you were then to feed a similar spreadsheet with one of the data from one of columns removed in the machine learning program, it will be able to fill in the missing values to a degree of accuracy. The accuracy of machine learning predictions are dependent on the amount of data that can be “learnt” from.

Predictive Analytics can be used to calculate the ROI of past marketing campaigns because of its ability to process large volumes of data and find patterns. Furthermore, once this has been done, further applications of the technology can be used to predict the ROI of future marketing campaigns.

Marketing will always have a large creative component, however with Predictive Analytics technology, marketing executives now have the option to bring scientific backing to validate their “gut feel”. With the ability to predict the ROI of marketing campaigns, marketing managers can potentially save millions of dollars without sacrificing on sales revenue.

Centazio is a new software solution that uses predictive analytics to calculate past marketing ROI and predict future marketing ROI. For those marketing managers looking for an edge on their competitors, it is well worth a look.

Tagged: #centazio #machine-learning #marketing #predictive-analytics #roi