The following visualization demonstates some simple and powerful capabilities of Tableau 5.1. It was published on Tableau's blog using the new free Tableau Public application, by Ross Perez. His entire blog entry is available here.
Subscribe To MIG Newsletter
Posts
All Comments
Wednesday, February 03, 2010
Monday, February 01, 2010
Veterans Health Administration Inpatient Satisfaction by Race
The Veterans' Health Administration health system is known for the consistent and high-level of care it is able to provide to American military veterans. Like any exemplary organization, it routinely monitors its performance in order to maintain very high standards for health outomes and patient satisfaction. Summary data on many of these recent efforts is available on Data.gov. Much of this information is summarized measures by VHA facility, which are organizaed in what the VHA calls VISNs. Each VISN may have one or more facilities under its umbrella.
The data is available in .csv format, making it easy to use in many analytical tools. For the following visualization we used the VHA_Facility_Quality_and_Safety_Report_Patient_Satisfaction.csv, which reports on overall inpatient satisfaction, stratified by race.
The Tableau visualization presents a dot plot of average satisfaction score reported for each facility in the dataset, with separate dots for each race. The facilities are listed in descending order by the average score reported for inpatient satisfaction among surveyed African-American patients. This relates to the major findings of the source survey, which found significantly different satisfaction ratings among African-Americans when compared to White patients and others. African-American patients were found to be generally less satisfied with their inpatient experience. The ordered list includes the state and VISN dimensions and for filtering on them as a way of drilling down to ask further questions, such as what is the distribution of satisfaction ratings within a state where more than one facility exists. The visualization makes it possible to quickly focus on one or more VISNs and to see whether satisfaction varied by state within a VISN.
The visualization also marks the average, and 2nd throiugh 3rd quartile ranges for the faciltiies included. Therefore, when the facilities list is filtered these measures change as well.
The accompanying report includes a detailed discussion of the findings on patient satisfaction, starting on page 15. It should help explain the context for the wide range of average patient satisfaction ratings and differences by race that this visualization highlights. The context includes the complexity of care offered at each facility, regional, cultural, and even faith-based norms and expectations of the served populations.
The data is available in .csv format, making it easy to use in many analytical tools. For the following visualization we used the VHA_Facility_Quality_and_Safety_Report_Patient_Satisfaction.csv, which reports on overall inpatient satisfaction, stratified by race.
The Tableau visualization presents a dot plot of average satisfaction score reported for each facility in the dataset, with separate dots for each race. The facilities are listed in descending order by the average score reported for inpatient satisfaction among surveyed African-American patients. This relates to the major findings of the source survey, which found significantly different satisfaction ratings among African-Americans when compared to White patients and others. African-American patients were found to be generally less satisfied with their inpatient experience. The ordered list includes the state and VISN dimensions and for filtering on them as a way of drilling down to ask further questions, such as what is the distribution of satisfaction ratings within a state where more than one facility exists. The visualization makes it possible to quickly focus on one or more VISNs and to see whether satisfaction varied by state within a VISN.
The visualization also marks the average, and 2nd throiugh 3rd quartile ranges for the faciltiies included. Therefore, when the facilities list is filtered these measures change as well.
The accompanying report includes a detailed discussion of the findings on patient satisfaction, starting on page 15. It should help explain the context for the wide range of average patient satisfaction ratings and differences by race that this visualization highlights. The context includes the complexity of care offered at each facility, regional, cultural, and even faith-based norms and expectations of the served populations.
Saturday, January 30, 2010
Testing dashboard image embedding
This example is provided by the Financial Research Station (FRS). It illustrates how live, interactive Tableau Server views can be served up on a public web site for persons logged into a Tableau Server with a valid user name and password. If you are not logged into the FRS Tableau Server, you will see a Tableau Server login screen below. The user name is frsbeta and the password is mkt2010. Contact FRS at 212-981-2529 x13 for assistance, or visit their web site for more details.
Friday, January 22, 2010
Employment and Payroll Data Example from Tableau Public
In an effort to explore the possibilities for using visual data analysis software to make public data transparent, exploratory, and understandable, we are testing the use of Tableau Software. One excellent and award-winning source of examples of public data is the Data Center of the Hamilton County Regional Planning Commission (HCRPC) web site. The resources there include data from local and national sourdes, including county departments; state governments in Ohio, Kentucky, and Indiana; and many federal agencies like the US Bureau of the Census and the Department of Labor (Bureau of Labor Statistics). The breadth of subjects include physical and policy planning, population change, school performance, economic indicators, and much more.
The following example takes data from the County Business Patterns, that was published in the County Business Patterns 2006-2007 .pdf by the HCRPC.
The map in the upper left quadrant of the faceted analytic display (Stephen Few, Dashboard Confusion Revisited, March 2007) shows the number of employees encoded as the size of each bubble, and the number of employees per establishment (color) for each county within the CMSA. Hovering over any bubble reveals the overall average number of employees per establishment along with the name of the county and the URL for the official county presence on the internet. Clicking on any bubble in the map causes the associated data in each of the other panels of the display to highlight.
The following example takes data from the County Business Patterns, that was published in the County Business Patterns 2006-2007 .pdf by the HCRPC.
The map in the upper left quadrant of the faceted analytic display (Stephen Few, Dashboard Confusion Revisited, March 2007) shows the number of employees encoded as the size of each bubble, and the number of employees per establishment (color) for each county within the CMSA. Hovering over any bubble reveals the overall average number of employees per establishment along with the name of the county and the URL for the official county presence on the internet. Clicking on any bubble in the map causes the associated data in each of the other panels of the display to highlight.
Wednesday, January 13, 2010
Wednesday, May 02, 2007
NoMail.us™ - the Easy Way to Opt-out of Direct Mail
Opting-out of direct mail cannot be easier with MIG's NoMail.us™ application for direct mail marketers.
Well-targeted and well-crafted direct mail should be based on database-driven analytics of best customers. When this is so, response and engagement/conversion rates go up and ROI shines. But not every best customer enjoys being marketed to directly for a variety of reasons. When this cannot be captured ahead of time (before prospect files reach the mail house), mail is returned at the marketer's expense and to the marketer's dismay.
Now, however, direct marketers can add some intelligence to their "do not mail" due diligence." MIG's NoMail.us service provides direct marketers with several ways to lower mail costs, study the self-reported opt-out group among their best customers, and increase response rates with more (better) qualified leads.
The NoMail.us service is a hosted, secure application MIG offers to marketers. It is used by CSRs, agency account supervisors and directors, marketing gurus and the like to manage the avalanche of returned direct mail and BRCs so the recipients are reliably removed from the prospect data stream of future campaigns. It is so simple to use, too.
And each client has their own NoMail.us database no one else can see.
Call 513-333-0345 to get started with your NoMail.us web site.
Thursday, April 26, 2007
Geography matters in business intelligence
Geography is a major part of business intelligence. As the folks at Environmental Systems Research Institute put it, "Geography Matters". In our work, "where" is often the first question. Who, what, why, how, and how much quickly follow, but "where" is usually central, along with the corollary, "how far?". For example,
Of course, who and how many are just as important. To get a start on these questions, visit www.DemographicsNow.com or check out the US Bureau of the Census population estimates page for a look at the Bureau's view of current population characteristics of the US and its geographic and political subdivisions.
- where to put the next store?
- where are the best customers or prospects?
- where are the competitors, and how far are they from your best performing stores?
- how far can the new sales territory boundaries be from the nearest distribution center?
- where are the high-value neighborhoods being under-served by your retail network
Of course, who and how many are just as important. To get a start on these questions, visit www.DemographicsNow.com or check out the US Bureau of the Census population estimates page for a look at the Bureau's view of current population characteristics of the US and its geographic and political subdivisions.
DataWell® Provides Hassle-Free Best in Class Marketing DB and CRM Services
Market Intelligence Group's best-in-class hosted marketing database and campaign management system, DataWell®, helps many companies solve customer relationship management challenges. The web-based solution delivers immediate comprehensive customer knowledge and powerful business intelligence directly to the desktop without IT overhead.
OVERVIEW
Customer knowledge allows your company to approach marketing challenges proactively and effectively. DataWell’s® flexibility allows users to meet users' changing marketing data management needs. The intuitive interface allows marketers to quickly and easily begin using the system without any need for complex commands, programming experience or in-depth data expertise. DataWell® enables informed marketing decision making.
MARKETING DATABASE
DataWell is a marketing database allowing your company to manage crucial data including business line transactional data, survey data, demographic data, and other data sources.
Any dataset that can be provided electronically can be incorporated into DataWell. Users can quickly set the scope of marketing campaigns in DataWell, and use flexible point-and-click and drop down selections to set conditional criteria for campaign targets, including current customers and/or prospects.
CAMPAIGN MANAGEMENT
Use DataWell to do more than creating data sets for analysis and mail lists. DataWell includes a full campaign management system allowing users to track and manage specific campaigns. Document business line, product, offer, target audience, quantity mailed, expenses and more for use later in calculating ROI on the campaign. Calendar functions allow you to easily see and track all campaigns currently in the market place.
CAMPAIGN TRACKING
The combination of the marketing database and campaign management modules allows DataWell to create automatic performance reporting based upon the purchase history of the targeted households within a defined response period. Use DataWell to show response rates, gross sales, and commissions, LTV, or whatever KPIs meet the business need.
Intuitive charts even show the response curves for a campaign, giving marketers a deeper understanding of the real response period for any given campaign.
REPORTING & STATISTICAL ANALYSIS
Access to the breadth of data in DataWell is a benefit for any company. The ability to organize, report and analyze your own customer data provides real return and insight.
OVERVIEW
Customer knowledge allows your company to approach marketing challenges proactively and effectively. DataWell’s® flexibility allows users to meet users' changing marketing data management needs. The intuitive interface allows marketers to quickly and easily begin using the system without any need for complex commands, programming experience or in-depth data expertise. DataWell® enables informed marketing decision making.
MARKETING DATABASE
DataWell is a marketing database allowing your company to manage crucial data including business line transactional data, survey data, demographic data, and other data sources.
Any dataset that can be provided electronically can be incorporated into DataWell. Users can quickly set the scope of marketing campaigns in DataWell, and use flexible point-and-click and drop down selections to set conditional criteria for campaign targets, including current customers and/or prospects.
CAMPAIGN MANAGEMENT
Use DataWell to do more than creating data sets for analysis and mail lists. DataWell includes a full campaign management system allowing users to track and manage specific campaigns. Document business line, product, offer, target audience, quantity mailed, expenses and more for use later in calculating ROI on the campaign. Calendar functions allow you to easily see and track all campaigns currently in the market place.
CAMPAIGN TRACKING
The combination of the marketing database and campaign management modules allows DataWell to create automatic performance reporting based upon the purchase history of the targeted households within a defined response period. Use DataWell to show response rates, gross sales, and commissions, LTV, or whatever KPIs meet the business need.
Intuitive charts even show the response curves for a campaign, giving marketers a deeper understanding of the real response period for any given campaign.
REPORTING & STATISTICAL ANALYSIS
Access to the breadth of data in DataWell is a benefit for any company. The ability to organize, report and analyze your own customer data provides real return and insight.
This is an Exciting Time for Business Intelligence
This is an exciting time for business intelligence. At least two developments offer enormous return on investment (and it does not have to be a huge investment). Web 2.0 technologies provide companies with opportunities to dive deeper into customer knowledge and to find the guidance they need for decision support. This can be quick e-mail surveys to test and tweak messaging before launching a campaign or gathering customer feedback on products and how they are used. New technology is allowing marketers to avoid mistakes and increase success. Also, data can now be quickly gathered and distributed via intelligent, Web-based dashboards. A large investment in people or hardware is not necessary to realize quick and actionable business intelligence.
The idea of this company is to bring all of the best tools and techniques for sophisticated data analysis and business intelligence to small and medium-sized businesses. Data mining is no longer done exclusively by the big companies. Market Intelligence Group has invested in developing the specific expertise, analysis tools and data visualization tools that enable small and medium-sized businesses to do the same.
As an analysis firm, we are using new media not necessarily to solicit the business intelligence but to deliver it. We are providing web-based tools that give instant access to business performance and intelligence down to the customer level—tracking what works and does not work. Generally speaking, all of our clients are embracing web-based tools that collect, organize and make sense of all of the data collected. Our clients want to focus on tools which communicate performance information directly, clearly, rapidly, and with proper context. They also want the information to be arranged in a way that logically supports their business process and focuses on what is currently happening vs. history.
While marketing has become more and more of a data-driven process, by no means has creativity or creative taken a back seat. In fact, being on-target with message is just as important as being on target with the audience; the message needs to resonate and communicate in a meaningful way. Getting noticed and creating share of mind with the correct prospects or customers in a highly fragmented media environment is a critical advantage for growing businesses. Research-driven marketing and better customer knowledge is now a necessary input to the creative process, and it supports relevant messaging. Knowing the buyer profile or their preferences goes a long way towards connecting.
The first question a company should be asking is: “What is the cost of failure?” When viewed within this context, business intelligence is generally a small investment overall; it can help avoid the failures and improve the chance for success for any businesss. In fact, business should be considered an investment that will build success, not an expense.
The idea of this company is to bring all of the best tools and techniques for sophisticated data analysis and business intelligence to small and medium-sized businesses. Data mining is no longer done exclusively by the big companies. Market Intelligence Group has invested in developing the specific expertise, analysis tools and data visualization tools that enable small and medium-sized businesses to do the same.
As an analysis firm, we are using new media not necessarily to solicit the business intelligence but to deliver it. We are providing web-based tools that give instant access to business performance and intelligence down to the customer level—tracking what works and does not work. Generally speaking, all of our clients are embracing web-based tools that collect, organize and make sense of all of the data collected. Our clients want to focus on tools which communicate performance information directly, clearly, rapidly, and with proper context. They also want the information to be arranged in a way that logically supports their business process and focuses on what is currently happening vs. history.
While marketing has become more and more of a data-driven process, by no means has creativity or creative taken a back seat. In fact, being on-target with message is just as important as being on target with the audience; the message needs to resonate and communicate in a meaningful way. Getting noticed and creating share of mind with the correct prospects or customers in a highly fragmented media environment is a critical advantage for growing businesses. Research-driven marketing and better customer knowledge is now a necessary input to the creative process, and it supports relevant messaging. Knowing the buyer profile or their preferences goes a long way towards connecting.
The first question a company should be asking is: “What is the cost of failure?” When viewed within this context, business intelligence is generally a small investment overall; it can help avoid the failures and improve the chance for success for any businesss. In fact, business should be considered an investment that will build success, not an expense.
Monday, November 22, 2004
PEAK: Flexible Reporting of Multi-Rater Employee Evaluations
PEAK Reports (Performance Evaluation And Knowledge) is our family of Excel and online applications for managing the production of multi-rater evaluation feedback reports. It offers outstanding time and process benefits to those involved in recruiting, screening, evaluating, managing, and developing employees. It frees HR coaches and organization development staff to concentrate on their primary tasks. The application provides for the production of several commonly used evaluation feedback reports, which can be customized to the needs of each client.
Behind the reporting engine is a multidimensional data cube based on data sets derived from any evaluation instrument. MIG processes this raw evaluation data using each client's business rules and evaluation objectives to produce the multidimensional PEAK data cube.
PEAK Reports adapt to a wide range of performance evaluation paradigms, and are not tied to a particular approach to performance evaluation. They accommodate any type of performance evaluation or employee screening process, including prescreening, interviewing, 90-, 180-, or 360-degree evaluations, gap reporting, skill assessments, etc.
The PEAK Reports reporting engine greatly reduces the time and effort that employment recruiters, HR staff, and other managers spend organizing evaluation activities, analyzing data, and preparing feedback reports.
The PEAK Reports application facilitates the quick turn-around of one-time, annual, or continuous reviews. It is available in turn-key, service-bureau, or blended configurations as a workstation, WWW, or server based tool, and requires no additional IT Investment. The output is hardware independent.
Because of its flexible structure, Peak Reports can be used for organizing and reporting on any activity or program for which consistent evaluative data is available. This might include treatment programs, standardized testing, benchmarking, evaluation of presentations and workshops, etc.
Behind the reporting engine is a multidimensional data cube based on data sets derived from any evaluation instrument. MIG processes this raw evaluation data using each client's business rules and evaluation objectives to produce the multidimensional PEAK data cube.
PEAK Reports adapt to a wide range of performance evaluation paradigms, and are not tied to a particular approach to performance evaluation. They accommodate any type of performance evaluation or employee screening process, including prescreening, interviewing, 90-, 180-, or 360-degree evaluations, gap reporting, skill assessments, etc.
The PEAK Reports reporting engine greatly reduces the time and effort that employment recruiters, HR staff, and other managers spend organizing evaluation activities, analyzing data, and preparing feedback reports.
The PEAK Reports application facilitates the quick turn-around of one-time, annual, or continuous reviews. It is available in turn-key, service-bureau, or blended configurations as a workstation, WWW, or server based tool, and requires no additional IT Investment. The output is hardware independent.
Because of its flexible structure, Peak Reports can be used for organizing and reporting on any activity or program for which consistent evaluative data is available. This might include treatment programs, standardized testing, benchmarking, evaluation of presentations and workshops, etc.
Monday, November 01, 2004
Maintaining Your Customers: Data Driven CRM
Customers today are very demanding and retailers must be able to anticipate customer needs or risk losing that customer.
For example: I grew up in a small town on the east coast of the United States. Every year before school started my mother would take me school shopping at the two or three local specialty stores in town. We knew the store owners by name and the owners knew our families. There was a sense of loyalty and the relationship between business and customer was a long standing one. If we did not find what we wanted or if we needed another size, the store clerk could place a special order and we waited, sometimes two weeks, for the product.
Today, back to school shopping is a very different process. We may go to one or two national retailers. We rarely know the store owners. Many times we shop online because it is possible to have access to regional, national or global retailers’ right in the privacy of our own home. Even though we do not know the merchants personally, we expect the retailers to understand and meet our needs. If we have to wait or if the retailer can not provide personalized customer service we will move on to the next merchant or supplier.
“We as customers want what we want and we want it now and at the right price.”
Fortunately today we have analytical tools and customer and product data to better serve and personalize the retail experience. Businesses know that it is less expensive to maintain a customer than to find a new customer. Therefore, customer relationship management (CRM) is key to a successful retail business. One CRM strategy that is gaining popularity is data mining.
Data mining is a process of knowledge discovery that employs statistical modeling to large amounts of transformed transactional and historical data, in an effort to understand and predict customer behavior.
The data utilized in the data mining process is typically massive and transactional in nature. It is opportunistic. The data was not collected with analysis in mind. Instead it was collected to monitor process control and track inventory. The data is usually messy and lacking integrity.
Right now ask yourself the following questions: Do you have access to large amounts of data? Is this data operational and/or transactional in nature?
If the answer is yes, you are lucky. You have the data needed to better understand your customers, hence making you a more responsive business partner. By uncovering patterns and relationships among your current customers, you will be able to put promotions and advertising in place that speaks directly to those customer segments. Once the power of your transactional data is unleashed you will be able to predict when a customer is about to churn. This will allow you to be proactive in the market place. Furthermore, you will be able to cross sell products and up sell products to existing customers.
You have the data what’s next? What is next is a process of data cleaning and data warehousing. You need to organize your transactional data so that you can build predictive models. The diagram below outlines the process. Stay tuned for the next edition of this newsletter when I will tell you how to household and prepare your data for the data mining process.
For example: I grew up in a small town on the east coast of the United States. Every year before school started my mother would take me school shopping at the two or three local specialty stores in town. We knew the store owners by name and the owners knew our families. There was a sense of loyalty and the relationship between business and customer was a long standing one. If we did not find what we wanted or if we needed another size, the store clerk could place a special order and we waited, sometimes two weeks, for the product.
Today, back to school shopping is a very different process. We may go to one or two national retailers. We rarely know the store owners. Many times we shop online because it is possible to have access to regional, national or global retailers’ right in the privacy of our own home. Even though we do not know the merchants personally, we expect the retailers to understand and meet our needs. If we have to wait or if the retailer can not provide personalized customer service we will move on to the next merchant or supplier.
“We as customers want what we want and we want it now and at the right price.”
Fortunately today we have analytical tools and customer and product data to better serve and personalize the retail experience. Businesses know that it is less expensive to maintain a customer than to find a new customer. Therefore, customer relationship management (CRM) is key to a successful retail business. One CRM strategy that is gaining popularity is data mining.
Data mining is a process of knowledge discovery that employs statistical modeling to large amounts of transformed transactional and historical data, in an effort to understand and predict customer behavior.
The data utilized in the data mining process is typically massive and transactional in nature. It is opportunistic. The data was not collected with analysis in mind. Instead it was collected to monitor process control and track inventory. The data is usually messy and lacking integrity.
Right now ask yourself the following questions: Do you have access to large amounts of data? Is this data operational and/or transactional in nature?
If the answer is yes, you are lucky. You have the data needed to better understand your customers, hence making you a more responsive business partner. By uncovering patterns and relationships among your current customers, you will be able to put promotions and advertising in place that speaks directly to those customer segments. Once the power of your transactional data is unleashed you will be able to predict when a customer is about to churn. This will allow you to be proactive in the market place. Furthermore, you will be able to cross sell products and up sell products to existing customers.
You have the data what’s next? What is next is a process of data cleaning and data warehousing. You need to organize your transactional data so that you can build predictive models. The diagram below outlines the process. Stay tuned for the next edition of this newsletter when I will tell you how to household and prepare your data for the data mining process.
Sunday, August 29, 2004
TURNING CUSTOMER DATA INTO KNOWLEDGE
MARKET INTELLIGENCE GROUP offers its clients the extensive experience of its principals in creating, managing, analyzing, and leveraging marketing databases for a number of national and global corporations, and not-for-profit organizations. The organization is comprised of :
senior level statisticians,
premiere web and desktop applications developers, and
business geographers.
Our staff has decades of experience in
senior level statisticians,
premiere web and desktop applications developers, and
business geographers.
Our staff has decades of experience in
business intelligence
database marketing
data mining
database administration
applied demographics
demand analysis
demographic forecasting
time series analysis and forecasting
business geography
MARKET INTELLIGENCE GROUP turns customer data into customer knowledge. We provide your business with marketing tools that facilitate positive customer interactions, improved sales performance and world-class customer satisfaction through more targeted, relevant, timely, and actionable decision-support.
Clients are in business-to-business and business-to-consumer markets:
Health care
Retail
Telecommunications
Education
Energy
Manufacturing
New technology
Entertainment/leisure
Financial services
Large government agencies
Not-for-profit organizations
Our solutions include:
DATABASE MARKETING
Data mining
Predictive modeling / target marketing
Segmentation strategies
Best Customer Profiling (using our BestProfile™)
RFM modeling
POS data analysis
Site selection and geostatistical analysis (using BestSite™ and state of the art GIS technology)
DATABASE DEVELOPMENT
List management
Merge-Purge capability / NCOA / CASS certification, etc.
Full campaign tracking (pre/post analysis)
CRM tools (DataWell™ and SalesLeader™)
RESOURCES/CAPABILITIES
Utilize sophisticated data analysis tools and techniques
CHAID • Logistic Regression • Cluster methods, etc.
Custom SAS programming
CONTACT US
Please feel free to contact us with any questions that you may have about our services and capabilities. We will respond to your questions as thorough and promptly as possible.
MARKET INTELLIGENCE GROUP, LLC
119 East Court Street
Cincinnati, Ohio 45202
voice: (513) 333-0345
Sunday, July 18, 2004
Subscribe to:
Comments (Atom)
