Marketing Analytics: Where to Start?

The first definition of marketing analytics that appears after a casual Google search is this:

Marketing analytics is the practice of measuring, managing and analyzing marketing performance to maximize its effectiveness and optimize return on investment (ROI).

Now, that’s easy enough to guess. Marketing Analytics is basically marketing professionals using analytics to help inform their work, and get more money. But how are these marketing professionals using analytics practically?

Well, after reviewing a few articles on the subject, the best breakdown I have found so far was on HBR.org:

This shows us various parts of the marketing cycle, and how much of a role analytics plays in each.

The Methods

Some of the methods they use to gather and analyze data include: text analysis (Idibon), machine learning (CrowdFlower), image analysis (AlchemyAPI), emotion analytics (Affectiva), and on the periphery: speech analytics, emoji analytics (Emogi). VentureBeat gives a quick overview of each of these.

Predictive Analytics

Predictive analytics are also very important to marketers.
By predicting what customers will buy in the future (i.e. Amazon suggestions, etc) or which customers will respond best to email coupons (i.e. Nordstrom Technology People Lab), predictive analytics give companies an edge on understanding what purchases will be made and how many customers they can, and will, attract. Simply put, it helps them save money on marketing. If you know who will react to what, then you can hone in on your target market, instead of sending out widespread (often useless) blasts. Although, these are best known for helping marketings working for e-commerce companies, according to Practical ECommerce. For a look at many of the smaller companies tackling this today, take a look at Custora, Everstring, and 6Sense.

One of the more audacious plans for predictive analytics would be Amazon’s latest patent: “anticipatory shipping.” This scheme would use marketing predictive analytics to pre-ship items that their algorithms assume you will likely purchase. Now, although they got the patent in 2014, they are finally reaching a day and age where that patent may become totally useful.

That’s enough for today, folks.
We’ll be back for more next week.
In the exciting world of marketing analytics.