Once upon a time, marketing experts relied on past experience and their own intuition in order to understand the customer. By using ex-post methods, such as focus groups and polls, a marketer would build an image of their customer and crunch limited available data to try to envisage how their product will position itself against the competition. This has changed immensely with services such as Facebook, Amazon and Netflix, which showed first-hand how data-driven marketing cannot just help understand the consumer but help to drive consumer behaviour and substantially increase gains.
But, how does this concern companies that do not have massive amounts of cash on hand for such activities?
There has never been a time before now where more data was available for analysis. If we just take a look at our time spent online, mobile users in the Philippines take the lead spending up to 10 hours a day in front of their screens. It’s easier than ever for marketers to follow digital bread crumbs to gather and analyse data on consumer behaviour which to influence the marketing strategies for their chosen niche.
The much-lauded leader in this field in the past couple of years has been Lenovo. According to them, they managed to develop a predictive model which helps them assert if a visitor to their website is going to buy one of their devices in a matter of seconds. By using this data, they carefully position customised content for the visitor with an accuracy of almost 90%.
Even though the times have changed, the founding principle of a data-driven marketing strategy is still a very old one – understand your consumer and their needs. There’s still no definitive answer on how to predict human behaviour. The rule of the thumb is finding data which can be analysed and segmented and thereafter expanded upon. For that reason, there are essentially three steps that need to be taken for a proper analysis to be conducted.
The first condition is to understand your consumer demographics, their interests, search analytics and how all these can be merged into forecasting consumer decisions. Using the Audiense tool provided by Meltwater, brands can search for keywords and terms to establish a consumer group that engages with that topic the most. Breaking it down by factors such as gender, age group and related interests marketers gain data-driven insights into who the audience is and an understanding of their behaviour as well as psychology looking into personality type and values.
After initial research, it’s important to divide your customer base into segments and prioritise target audiences who have the most interest in your field, particularly consumers with intent to buy. Using the Audiense tool, you receive analytics on consumer behaviour and traits based on a specific discussion point related to your brand. This can then help influence marketing and advertising decisions.
In order to have a complete overview of your market, competition analysis is essential. Add to that the analysis of consumer behaviour online and a company can be said to be prepared for connecting with and understanding their target audience properly.
Data-driven marketing is thus grounded in creating the image of the perfect consumer based on real people being analysed, thus focusing on influencing their decision-making process and attempting to predict their behaviour over time.
Once you have your analytics on consumer behaviour, the question that is naturally posed is – how can you actually predict anything?
We all buy certain items at certain times in our lives, the so-called life stage based purchases. Every time we buy a new car, a home, marketers can use this information to predict complementary purchases. By using these analytics effectively, they can target consumers by creating helpful content marketing such as articles or recommendations to increase sales.
This still might seem like some kind of dark art, but it is actually entirely based on what we do. Even though it is not possible to look for “people who just bought a new car”, it is possible to analyse existing databases, see people who have, at some point, bought a product and backtrack the behaviour that led to the purchase. By correlating this data with age, gender and income, it’s much easier to segment the market and find connections, consequently making much more successful predictions for the future.
This is the trickiest part of data-driven marketing. It’s much easier to understand consumer behaviour and interests of target demographics than the intent to buy. It is far different to just know that someone will buy something than knowing when that person is going shopping
This is where connecting databases is essential for marketers to devise a strategy. If a company is able to create a data partnership and gather information on shoppers from other websites and services, it will have an easier time measuring consumers’ intent. The universal wisdom here is as follows – if a person is looking at articles about cars, we can safely assume that the person is interested in cars. But, if that same person clicks on product pages of an actual used and new car lots, this is a much better indicator of intent to buy.
If you have a system set up, the information derived from it about your customers and their habits is a virtual El Dorado, limited only by your imagination and the skills of your data scientist. For example, a company with a telephone answering service which can correlate their customer insight at the door, so to speak, can achieve an incredible boost to their business.
The implications are endless, whether they are connected with empowering your sales department by giving them hot points which will result in a buy or your R&D by giving them insights into what your customers expect from your product.
Data harvesting is the new way of doing business, period.
About the author: