Big Data and Customer Targeting

Facebook, LinkedIn and other social media and networking websites coupled with smartphone GPS, AI and big data analytics allows the marketer to target customers in a way that was never possible before. 

Not only can the marketer target based on demographics, psychographics and physical location but also on intangibles such as highly correlated behaviors, purchases, web searches, affiliations, etc. 

So first, through factor analysis, cluster analysis and other big data analytics, you need to discover what correlates most closely with the purchase of your brand. Once you have discovered this, then you need to find ways to target people with those attributes. This allows for hyper-efficient targeting that makes old fashioned media planning seem quaint. 

Sometimes it is as easy as using the Facebook platform to run a highly targeted ad. At other times, you have to be creative about how to reach people with certain attributes.  

The beauty of this approach is that you can send very specific messages to each target group or segment. Each message will focus on the angle that will most appeal to people in that group or segment. 

Microtargeting is a related concept. it is the use by political parties and election campaigns of direct marketing datamining techniques that involve predictive market segmentation (aka cluster analysis). It is used by United States Republican and Democratic political parties and candidates to track individual voters and identify potential supporters. The term "microtargeting" was coined in 2002 by political consultant Alexander P. Gage.

When you combine prospect location with targeted messages on mobile phones you are entering the realm of geo-fencing and geo-targeting. Because I know A and B about you and because I know that correlates with an interest in my product and brand and because I know you are in a physical location proximate to where my brand is available for purchase, I will send you a highly personalized alert using the most compelling message for you indicating that I will give you an incentive to purchase my brand now. For an example of this, click here.

While brands need to be managed at a global level, increasingly they need to be marketed at an individual level. With today's tools, that is becoming more and more possible. 

Big Data and the Evolving Skill Set of Marketers

With the advent of social media and big data analytics, an amazing amount of information has now been amassed on most consumers. The trick is to know how to use that data to achieve organizational goals. Marketers' standing in organizations has the potential to increase significantly if they are able to successfully tap into this information to achieve these organizational goals. This starts with understanding who an organization's best customers are - those who are the most loyal and the most profitable and who deliver the largest lifetime value. This requires analysis of sales and profitability data against demographic, psychographic, geographic and other customer data to identify individuals and highly targeted market segments who promise the greatest ROI for the organization in question. Big data analytics can also identify other less intuitive but no less predictive indicators of high loyalty, profitability and lifetime value. 

The marketer must then know how to tap into a wide variety of databases to identify these people and send targeted, personalized messages, offers and other invitations to them. 

If the marketer is able to connect all of these dots, he or she is certain to become invaluable to his or her organization.

Big Data Analytics

This is the process of analyzing large amounts of unstructured or semistructured data of various types across different sources to identify patterns and correlations that could provide better targeting, additional revenue opportunities, and other competitive advantages. Some uses of big data include:

• More in-depth and precise business understanding
• Improved customer relationship management (CRM) campaigns
• Optimized segmenting of customers
• Improved market trends and analysis
• Recognition and development of sales and market opportunities
• Improved planning and forecasting

Big data analytics is enabled primarily by substantially improved statistical and computational methods, but also by improved storage and computational capacity. Rather than constructing the experiments first, big data allows one to accumulate vast quantities of data and then identify statistically significant patterns. The largest challenge to big data analytics is an internal skills gap. Big data can reveal unexpected correlations. My favorite is that the purchase

of furniture coasters is a strong indicator of low credit risk and high credit scores.

Jason Frand of UCLA Anderson Graduate School of Management provides these examples of how big data is being used today:

• "One Midwest grocery chain used the data mining capacity of Oracle software to analyze local buying patterns. They discovered that when men bought diapers on Thursdays and Saturdays, they also tended to buy beer. Further analysis showed that these shoppers typically did their weekly grocery shopping on Saturdays. On Thursdays, however, they only bought a few items. The retailer concluded that they purchased the beer to have it available for the upcoming weekend. The grocery chain could use this newly discovered information in various ways to increase revenue. For example, they could move the beer display closer to the diaper display. And, they could make sure beer and diapers were sold at full price on Thursdays. . . .
• American Express can suggest products to its cardholders based on analysis of their monthly expenditures.
• WalMart is pioneering massive data mining to transform its supplier relationships. WalMart captures point-of-sale transactions from over 2,900 stores in 6 countries and continuously transmits this data to its massive 7.5 terabyte Teradata data warehouse. WalMart allows more than 3,500 suppliers to access data on their products and perform data analyses. These suppliers use this data to identify customer buying patterns at the store display level. They use this information to manage local store inventory and identify new merchandising opportunities. In 1995, WalMart computers processed over one million complex data queries. In 2010, WalMart processed 1 million customer transactions every hour feeding 2.5 petabyte databases.
• The National Basketball Association (NBA) is exploring a data mining application that can be used in conjunction with image recordings of basketball games. The Advanced Scout software analyzes the movements of players to help coaches orchestrate plays and strategies. For example, an analysis of the play-by-play sheet of the game played between the New York Knicks and the Cleveland Cavaliers on January 6, 1995 reveals that when Mark Price played the guard position, John Williams attempted four jump shots and made each one! Advanced Scout not only finds this pattern, but explains that it is interesting because it differs considerably from the average shooting percentage of 49.30 percent for the Cavaliers during that game."

As additional examples:

• Netflix mines its customer digital download database to recommend movies and shows that their customers may like, based on their viewing history.
• Xerox is developing social media analytics software to enable businesses to monitor their brand images. The software will be able to identify specific themes in social media content (tweets, blog posts, etc.) and route them to the appropriate internal people to handle.

Excerpted from Brand Aid, second edition. Order your copy here now.

The Future of Technology in Marketing

Between Google and Facebook, your every online move is monitored. If you take online quizzes, those can provide insight into your personality and motivations and even your IQ. Through GPS, your mobile device can follow your every movement. Security cameras capture you or your automobile and its license plate wherever you go. New wearable devices monitor your blood pressure, heart rate, breathing and skin moisture and therefore your state of mind. Apps can measure the number of paces you have taken in a day. The camera on your Google Glass can capture what you are seeing. And big data analytics can discover important patterns and correlations between data sources.

We are truly entering a world in which almost everything can be known about us. How might this play out in marketing? What if marketers could figure out not only what we had bought in the past but also what we might buy in the future and what if they could link that to geotargeting (location-based marketing) and geofencing (combining location and timing in marketing messaging)? And what if, knowing your circle of friends and who you are closest to through social media, they can tell you which friend just bought something similar?

For example, you have been thinking about buying a pair of Ferragamo shoes and you walk by a boutique that carries the pair of shoes you have been considering purchasing. Your mobile device vibrates in your pocket and alerts you to the fact that you are in front of a store that has what you want and it offers a time-sensitive discount on the shoes if purchased at that store. Further, it mentions that your best friend bought a pair of Ferragamo shoes last week (peer pressure/reinforcement). Upon detecting elevated heart rate and more rapid breathing, your personal device prompts you again with another selling message for those shoes. Far fetched? Not really. Welcome to the 21^st Century.