Join the dots: Know thy customer in the era of big data
As a kid, I loved join the dots puzzles. The problem was that the puzzles with lots of dots held no surprise; the picture was usually clear for all to see. Big data is like join the dot puzzles: the more dots you can populate, the easier it is see the picture.
As customers demand more personalized, targeted communications, companies have to exploit all means at their disposal to understand and deliver on their needs. Connecting the dots of big data is one tool companies must exploit.
To successfully use big data to improve the customer experience, companies must do three things.
- The first is integration of data sets. To return to my join the dots metaphor, imagine how difficult the puzzle would be if the dots for one picture were in several different books: sales data in one book, process measures in another, and feedback in a third. This data fragmentation is the first challenge many companies face in exploiting big data. It is why Clicktools puts integration of all customer information into CRM – the system of record – at the heart of our technology. Bringing together customer, operational, sales, and feedback data from all stages of the customer journey is step one.
- Whilst many commentators focus on manipulating massive datasets, they overlook the real value of how different datasets inter-relate. So-called linkage analysis is based on insightful questions like “Which attributes of customer experience do high-spenders think are most important?” Or, “Satisfaction with which product attributes lead to the lowest churn of the different customer segments we serve?” Or, “What is the lifetime value of our most and least satisfied customers?”Figuring out which questions to ask is a key skill for successfully exploiting big data. As quality guru W. Edward Deming said, “If you do not know how to ask the right question you discover nothing.”One other skill to master is understanding the difference between correlation and causation. Statistics may suggest that two datasets are highly correlated (i.e. the size of one dataset reflects the size of another). This relationship, however, does not mean one causes the other. For example, research from a UK phone provider showed a very high statistical correlation between phone usage and cat ownership, but no one has ever been able to explain why cat owners spend more time on the phone.
- Action is the final and most important piece of the big data puzzle. Guiding improvements is the only reason for investing in big data, so if you are not prepared to act, don’t waste time and money in any analysis. Analysis without action is for academics, not businesses. Successful business man Thomas Edison expressed this quite well, saying, “The value of an idea lies in the using of it.”
Finally, in all the big data hype, let’s not lose sight of little data, the individual dots of customer interaction that make up the big data puzzle. Without each of these, there would be no big data and if ignored, there might be no business!