CH 5
These operations are so complex that often companies need to build a data warehouse
(which can cost more than $10 million) simply to store and process the data.
Cost-per-impression
-Advertiser is charged each time ad shows up on user page -Less expensive, but not as easy to measure in which the cost of the advertisement is charged each time the advertisement shows up on a page that the user views.
Cost-per-click
-Advertiser is charged only when user clicks on ad -More expensive, requires greater interaction in which the cost of the advertisement is charged only each time an individual clicks on the advertisement and is directed to the Web page that the marketer placed within the advertisement. This method of charging for advertisements is common for online vendors of advertisement space.
Unstructured Data
-Body of emails -Tweets -Facebook Status Update Messages -Video Transcripts (harder to get)
Commercial Entities Many companies today collect data in large quantities to sell to other organizations
-Credit card purchase data -Supermarket scanner data
Uses of Data Mining
-Customer Acquisition -Customer Retention -Customer Abandonment -Market Basket Analysis
Structured Data
-Date -Time -Census Data -Facebook "Likes" (easy to get)
Millions of pieces of information that make up Big Data originate from two source categories:
-Direct path (You shop for a car online and see a model that you like. You submit a request for information form in which you supply personal information, including features of the car that appeal to you. That information is stored in the car dealership's database, and a salesperson pulls it up later on before she contacts you about the car.) -Indirect path (On the other hand, data creation can be a by-product of another action. A company that uses Big Data might know, for example, that consumers who purchase green detergent products, register as Democrats, and hold college degrees are more likely than average to purchase a hybrid vehicle. A person who fits this profile might receive a communication about a Honda Prius or other hybrid car even though he or she has not (yet) specifically requested information about these vehicles.)
One-to-one marketing includes several steps
-Identify customers and get to know them in as much detail as possible -Differentiate among these customers in terms of both their needs and their value to the company -Interact with customers and find ways to improve cost efficiency and the effectiveness of the interaction -Customize some aspect of goods or services offered to each customer
Big Data can provide competitive advantages in three main areas:
-Identifying new opportunities through analytics that yield greater return on marketing investment -Transforming insights into better products and services that are better aligned with desires of consumers -Delivering timely information more efficiently (Delivering communications on products and services to the marketplace more efficiently and effectively) These insights can enhance firm profits, but may also benefit society as a whole
Characteristics of CRM
-Lifetime value -Customer equity -Customer prioritization -Share of Customer
Data analysts have traditionally focused on structured data
-More readily obtainable -Computers today can easily analyze a large number of data points.
Partner Databases provides benefits to buyers and sellers
-Real-time demand signals -Replace inventory with information -Fewer stock-outs
CRM leverages database technologies to customize customer interactions based on:
-Share of customer -Lifetime value -Customer equity -Customer prioritization
Digital Marketing Channels
-Social Media -Pay per Click Advertisements -Search Engine -Email
Increased types and amounts of government-generated data are accessible to enterprising marketers. (Provided by the government and non-governmental organizations.)
-U.S. Census -Index of Economic Freedom -Bureau of Transportation Statistics -Recalls.gov
Examples of CRM in Action
-USAA (always talk to a USAA representative) -Amazon.com (tracks visits) -Disney's My Magic+ (wearable Magic Band)
Social Media Sources
-Web scraping -Sentiment analysis -Measuring brand attitude by assessing the context or emotion of online comments -Brand mapping
Market basket analysis
Develops focused promotional strategies based on the records of which customers have bought certain products. Hewlett-Packard, for example, carefully analyzes which of its customers recently bought new printers and targets them to receive e-mails about specials on ink cartridges and tips to get the most out of their machines.
Up to now, discussion of marketing analytics has focused on validating prior investments
Focus on understanding current performance
Four steps of one-to-one marketing
identify - collect and enter names and additional information about your customers. Differentiate - Identify top customers. Interact - Call the top three people in the top 5 percent of dealers, distributors, and retailers that carry your product and make sure they're happy. Customize - Find out what your customers want.
CRM systems enable marketers to
identify priority customers and customize communications and special offers accordingly-Personal selling-high volume, direct mail-low volume
Big Data can easily exacerbate the problem of information overload,
in which the marketer is buried in so much data that it becomes nearly paralyzing to decide which of it provides useful information and which does not.
Many firms look to increase share of customer,
instead of share of market.
Customer relationship management (CRM)
involves systematic tracking of consumers preferences and behaviors over time in order to tailor individualized value propositions -Allows firms to get "up close and personal" -A process by which firms enact their customer orientation -Capture information at each customer touchpoint
Big data
is a popular term to describe the exponential growth of structured and unstructured data -Internet data can be hard to analyze using traditional approaches -Internet of Things
The Internet of Things
is a term that is increasingly used in articles and stories on technology trends to describe how everyday objects (e.g., cars, refrigerators, etc.) are connected to the Internet and in turn are able to communicate information throughout an interconnected system.14 Areas that would become part of this network include medical devices, cars, toys, video games—the list goes on and on.
CRM firms view customers differently - as assets. Customer equity
is defined as the financial (net) value of a customer throughout the lifetime of the relationship.
Customer equity
is financial value of a customer relationship-Bank -Takes into account monetary investments to acquire and maintain relationship
Lifetime value of a customer
is how much profit a firm will make on a customer
The final characteristic that makes CRM unique
is the fact that organizations focus on high-value customers. This means the firm prioritizes its customers and customizes communications accordingly.
Share of customer
is the percentage of a given customer's purchases in a category over time -Enables company to grow sales and profits at a lower cost, relative to new customer acquisition
Sources of Big Data
- Government and NGO - Commercial Entities - Partners - Social Media - Corporate IT Sources These sources can be both within and outside of the organization and created and compiled from different groups
1 terabyte =
1,024 gigabytes
Acxiom acknowledges it has on average
1,500 pieces of information on 200 million+ Americans
in a CRM world
80% of the profits often come from 20% of the customers.
Share of customer
Because it is always easier and less expensive to keep an existing customer than to get a new one, CRM firms try to increase their share of customer, not share of market; this is the percentage of an individual customer's purchase of a product over time that is the same brand. For this reason, CRM focuses on increasing a brand's share of customer.
Lifetime value of the customer
Lifetime value is the potential profit a single customer's purchase of a firm's products generates over the customer's lifetime. Estimating LVC requires that the firm first estimate future sales across all products for the next 20 or 30 years, and then that the firm attempt to figure out what profit the company could make from this customer in the future.
Customer aquisition
Many firms include demographic and other information about customers in their database. For example, a number of supermarkets offer weekly special price discounts for store "members." These stores' membership application forms require that customers indicate their age, family size, address, and so on. With this information, the supermarket determines which of its current customers respond best to specific offers and then sends the same offers to noncustomers who share the same demographic characteristics.
Deriving meaning from unstructured data is more difficult, but potentially more valuable
New technologies are making this process easier
Customer Prioritization
Not all customers are equal ... at least, not in terms of profitability!
Customer abandoment
Strange as it may sound, sometimes a firm wants customers to take their business elsewhere because servicing them actually costs the firm too much. Today, this is popularly called "firing a customer." For example, a department store may use data mining to identify unprofitable customers—those who don't spend enough or who return most of what they buy. For example, data mining has allowed Sprint to famously identify its customers as "the good, the bad, and the ugly."
Data Mining
The biggest data challenge for many firms is determining what to do with it all!
Customer retention and loyalty
The firm identifies big-spending customers and then targets them for special offers and inducements other customers won't receive. Keeping the most profitable customers coming back is a great way to build business success because—here we go again!—keeping good customers is less expensive than constantly finding new ones.
Partner Databases
Two-way information exchange between purchasing organization and suppliers
According to SAS,
a leading provider of data analytics software, "Big Data refers to the ever-increasing volume, velocity, variety, variability and complexity of information."
Data sold in
aggregate form
Sentiment analysis
analysis is a process by which analysts seek to identify changes in customers' attitude toward a brand by assessing the context or emotion of comments provided.
Each action you take online leaves a digital imprint,
and all of those imprints have the potential to yield valuable insights for a wide range of stakeholders within society.
Data brokers
are companies that collect and sell personal information about consumers, including: -Religion and ethnicity -User names -Income -Medication they take, and more ...
Structured data
are data that (1) are typically numeric or categorical; (2) can be organized and formatted in way that is easy for computers to read, organize, and understand; and (3) can be inserted into a database in a seamless fashion.
Unstructured data
are often qualitative in nature and do not possess all three of these properties. contain nonnumeric information that is typically formatted in way that is meant for human eyes and is not easily understood by computers. ex. body of an e-mail message
Scanner data
data derived from all those items that are scanned at the cash register when you check out with your loyalty card (which just happens to have your demographic profile information in its record!).
CRM
facilitates one-to-one marketing
Reality Mining
mining is the collection and analysis of machine-sensed environmental data pertaining to human social behavior with the goal of identifying predictable patterns of behavior. Reality mining was declared to be one of the "10 technologies most likely to change the way we live" by Technology Review Magazine.
Commercial Entities May be a
primary or secondary source of revenue for the firm
Data mining
refers to process by which analysts sift through Big Data to identify unique patters of behavior -Data warehouses -Data brokers -Reality mining
Commercial Entities Data sold in aggregate form
so that it's not possible to identify the actions of a specific consumer, but scanner data still provide extremely useful information to both manufacturers and retailers about how much shoppers buy in different categories and which brands they choose.
Data brokers
specialize in collecting and selling personal information about consumers.
Data derived from data mining efforts can be broadly classified as
structured and unstructured data
As new technologies continue to enhance the ways we connect to people, machines, and organizations,
the volume and complexity of Big Data will continue to increase.
Its easier and less expensive
to keep a current customer than it is to acquire a new one.
Predictive analytics
use large quantities of data to more accurately predict future outcomes
Web scraping
using computer software known as web crawlers to extract large amounts of data from websites.