Decoding Data Science For Digital Marketers

Decoding Data Science For Digital Marketers

The exponential growth of data is transforming digital marketing. According to Booz & Company, 61% of data professionals say big data will overhaul marketing for the better. A stunning 90% of all data has been generated over the last two years (source: SINTEF). The vast majority of this data is generated from social media and is “unstructured data.” The availability of affordable high-speed computing power and the emergence of software like Hadoop and Spark are democratizing the ability to synthesize large amounts of data previously only available for large institutions and enterprise companies.

The application of data science allows digital marketers to develop actionable conclusions to affect positive business outcomes. Marketers like myself are equally intimidated and fascinated by data science. And marketers love to throw around terms like big data, algorithms and predictive analytics without fully understanding the origins or the original intentions of the academic terms.

To help get digital marketers to start to speak data science, I selected the top three topics to make you look smart in the war room:

1) Defining “Big Data”

Marketer: Any dataset stat that can’t fit on a spreadsheet i.e the transition from Excel to Access.

Academic: There is no threshold that defines how much data defines “big data”. The assumption is that analytics is inherent in big data.

How to use “big data” in conversation:
Setting: Cocktail party
“Little data can be just as valuable as big data, it’s just how you use it” (pause, wait for the slap to the face). And scene.

2) Unstructured vs Structured Data

Marketer: Wait. there’s two different types of data? Who knew?

Academic: “Unstructured” data means data that is spontaneously generated and not easily captured and classified. “Structured” data is more akin to data entered into a form, like a user name might be, or generated as part of a pre-classified series, like the time stamp on a tweet.

Examples of data:

Social media: The vast majority of social media data is unstructured data, i.e. tweet or a blog. metadata is structured
CRM: structured except for comments
Mobile: Meta data on phone actions is structured data, SMS/text messages are unstructured
Transactional (Credit card / POS): structured
Weather: structured

3) Machine Learning Versus Heuristic or Rule-Based Systems

Marketer: The vast majority of enterprise social media platforms have analytics that can help me track word counts and sentiment on keywords. I’m not too happy with the sentiment and the ability to find actionable insights on simple summations of followers, likes and retweets.

Academic: Machine learning is a subfield of computer science and statistics that deals with the construction and study of systems that can learn from data, rather than follow only explicitly programmed instructions like the first-generation and antiquated social media platforms.

Example: Identifying parent profiles

In a heuristic or rule based system, a user will setup keywords like “mom”, “dad”, “father”, or “mother” to identify conversations or profiles that use those words. Since it is a rigid system looking only for exact matches on user inputted keywords, the heuristic system will miss users who might not self identify themselves as a parent in their profile using one of the keywords but may talk about “dropping my kids off at school”. A machine learning system can use conversations like “dropping my kids off at school” to identify those users as parents. The next generation digital marketing platforms will us machine learning to help marketers and advertisers glean insights from this vast ocean of unstructured consumer data collected by the world’s largest social networks.

There are stark differences in style, terminology and problem solving methodologies between the digital marketing and academic worlds. Nevertheless, if you can successfully find the right mix of the two, it provides insights that allows sophisticated digital marketers to create braver, bolder more personalized content for their audience. Audience targeting and personalized predictive marketing using social data are expected to be some of the business areas that benefit the most from mining big data.

To see personalized predictive marketing in action, request a demo of the People Pattern platform.