What Are You Doing with Your Data?

Do you have a list of prior customers or vendors? How do you compile that information, and do you use it to benefit your business? In this article, Linda begins to look at how data mining can help increase your sales and services.

Clarifying Data Mining Objectives

Large-scale enterprises use data mining extraction to connect transaction and analytical systems. In other words, when a grocery store uses data mining to determine sales, they may find that more men than women buy diapers on Thursdays, but that beer sales also escalate on Thursdays. This might mean that women are preparing for the weekend, sending husbands out for errands. While out, the husband buys beer for whatever physical or psychological reason at the same time as they buy the diapers for the babies.

What that store does with that information depends upon how relationships are analysed.  They can use data mining software to analyse those relationships and patterns. Several types of analytical software are available: statistical, machine learning, and neural networks. Generally, any of four types of relationships are sought:

• Classes: Stored data is used to locate data in predetermined groups. For example, a restaurant chain could mine customer purchase data to determine when customers visit and what they typically order. This information could be used to increase traffic by having daily specials.

• Clusters: Data items are grouped according to logical relationships or consumer preferences. For example, data can be mined to identify market segments or consumer affinities.

• Associations: Data can be mined to identify associations. The beer-diaper example is an example of associative mining.

• Sequential patterns: Data is mined to anticipate behaviour patterns and trends. For example, an outdoor equipment retailer could predict the likelihood of a backpack being purchased based on a consumer's purchase of sleeping bags and hiking shoes.

Just to go one step further (as this may make sense to visual communicators), data mining consists of five major elements:

• Extract, transform, and load transaction data onto the data warehouse system.

• Store and manage the data in a multidimensional database system.

• Provide data access to business analysts and information technology professionals.

• Analyse the data by application software.

• Present the data in a useful format, such as a graph or table.

"Ah-ha!" you might exclaim. Visual imaging! Information graphics, or infographics, are the end result of many of these analysing efforts:

Every image shown above was created from analysed databases. That information was simplified and made visual, or "pretty," so people who needed to learn from that data could easily read it. So, if you've created infographics in the past, or if you create them now, you are part of that analysing process.

This same process is computerized, and an example is when you view maps and diagrams at Google Analytics or on your Facebook Fan Page's Insights. In fact, it doesn't take much analytical savvy to realize that the following information can offer some clues about readers and possible shoppers:

The information shown above displays the gender and age of a Fan Page at Facebook that focuses on a group of retail businesses in a small town. It doesn't take a rocket scientist to understand that the largest readership is between ages 35-44, and that most of all readers at any age are female (except the 13-17 year-old males, which is curious to me).

But, what do I do with that data? Do I want to increase male readership? If so, why? Do I want to cultivate and increase the female readership from ages 44 and up? If so, then why?

At this point, I realize I need more data to inform any further progress. All I have at this point is classes of information (from the previous list). What I want to do now is cluster that information with other statistics that might tell me why this page's readership is classed as it currently stands. And, I can begin to make associations as well...

For instance, I might want to learn if these individuals are shopping at the stores represented by this Facebook Fan Page. That information is not available through Facebook. Therefore, I might need to create a survey. In that survey, I would ask for ages and gender...so I can make associations with the data shown above.

Then, I can continue those surveys to make sequential patterns. But, at that point, I might realize that I need to save data from previous weeks, months or years for comparison. While I could collect this data and analyse it manually, I don't have that amount of time on my hands. Instead, I would use software where data could be input and where comparisons and sequential patterns could be spit out at my convenience.

Stay tuned for that information, upcoming in the next article...