Leverage Big Data
When asked the question: what are the most important issues the marketing industry is facing today? I consider the most important issue is how to leverage big data in the proper way to drive revenue and guide strategy.
Technologies bring convenience to marketers especially on collecting data during any customer’s tiny movements towards a decision. The purchase funnel is narrowed down through big data---customers have been “thin sliced”, and clustered into different segmentations. Marketers consider big data as their savior on strategy making. They rely on data and think data never lies. I also believe data is one of the great milestone in marketing. I am also fully impressed by the power of data especially when I am running the cluster analysis results in SAS and reading the correlation between different factors. However, does data really tell the true story or does it distort the truth? The double-edged sword of data is that it can be easy to optimize locally but fail globally (and vice versa).
As one of the effective ways of machine learning, data can only give a basic reference to support truth (it is not always telling the real truth). Understanding the customer’s unmet needs is still considered a challenge. For example, your customer makes an alternative option on purchasing Product B because the Product A that they desire is out of stock. In this circumstance, the data would say the customer’s preference is Product B instead of their real preference, Product A. You get a solid number, but at the same time, you miss opportunities by misinterpreting their behavior. It happens in both B2B and B2C marketing.
Smart marketers rely on predictive technologies that do the analysis for them and suggest where to make their best marketing bets. Data can help find answers to quantitative questions like which account contributes the most business, and which accounts sales and marketing should be gone after. Data can't solve everything, but the best marketers fuse intuition and analytics to make bold moves, measure, and iterate.
In my point of view, data should never be read separately, but should be put into the right context to interpret. Considering the big data trend in the current marketing industry, the most terrible things I consider is not data invisibility but the misunderstanding and interpretation data.
