If you’re a media student or simply anyone who pays cursory attention to current happenings, chances are you’ll be sick to death of hearing about ‘big data’, ‘machine learning’ and ‘neural networks’. It stands to reason, however, that your understanding of these concepts and their implications for marketing, media and communications is a little less comprehensive than your contempt for these oft bandied terms. This article aims to explain these concepts in an easy to understand manner, and explain what they mean for the future of the industry.
Big Data: a nice, cover-all term for extremely large datasets that are analysed to discover patterns and trends. A great example of big data in use is Facebook. Ever wonder why you get ads for an extremely specific product, brand or business that suits your needs or desires just perfectly, despite never having mentioned this desire anywhere near Facebook? That’s because Facebook has over a billion users, and is able to create extremely accurate consumer models through the data gleaned from their users. Other Facebook users out there are very similar to you in terms of likes, dislikes, shopping habits, demographic, psychographic and behavioural elements, and, put simply, Facebook guesses you’ll like some of the things they like as well. That’s big data – as the name suggests, it’s just a lot of data.
Machine Learning: a type of Artificial Intelligence that allows computers to learn new information without being explicitly programmed to, changing and learning in response to new data. Yeah, it’s exactly as science-fictionesque as it sounds. Machine learning is how big data is analysed, allowing for complex relationships between variables to be formed. In more simple terms, machine learning is a computer finding patterns in data and then being able to accurately predict other patterns and relationships. Machine learning is how the aforementioned analytical and consumer models are formed.
Neural Networks: an analytical tool that imitates the neural network of the human brain, attempting to solve problems in a similar manner (just much faster). Neural networks are unique because they do not require statistical formulas or variables that are considered important before running. Additionally, they can find hugely complex relationships between any number of seemingly unrelated variables. Because they model the human brain, Neural networks are also especially good at dealing with imperfect and un-sanitised data. Relating it to the previous two concepts, machine learning is the process that neural networks use when analysing big data.
Ok, but how does this relate to me?
Good point. Pattern recognition and machine learning is THE emerging frontier in marketing, and as an extension, media and communications. As you were reading this, the cogs should have been ticking away in your head; the implications of a method that can extremely precisely find the perfect target market for a brand or product should be rather glaringly obvious, but if not I’ll spell it out for you.
- Product and brand experience can be individually tailored to each customer, including both pricing and marketing strategies
- Highly precise customer segmentation, creating extremely homogenous target markets
- Accurate prediction of customer lifetime value
- Rich analysis of highly complex data to form deeper insights than previously possible
These are just a few of the advantages of effectively employing big data in marketing strategy.
The bottom line
What it does: turns complex datasets with a large number of variables into easily interpretable and implementable information for use in marketing and analytics
Why this matters for you: If you’re reading this, that means you’re probably a student in media, communications and marketing, or already employed in the industry. If you’re the latter, you’re probably way ahead of the curve here, but if you’re the former this is the sort of cutting edge techniques you’ll need to learn to master in order to succeed in the industry in this day and age. Big data turns numbers into something actually interpretable and useful, something that’s important regardless if you’re a marketer, journalist, business owner, radio host or one of a host of new professions being created on the daily thanks to frontier technologies like big data.
By Tom Yarnold