Portals eNewsletters Web Seminars dataWarehouse.com DM Review Magazine
DM Review | Covering Business Intelligence, Integration & Analytics
   Covering Business Intelligence, Integration & Analytics Advanced Search
advertisement

Resource Portals
Business Intelligence
Business Performance Management
CRM
Data Integration
Data Quality
Data Warehousing Basics
EAI
EDM
EII
ETL
More Portals...

Advertisement

Information Center
DM Review Home
Web Seminars & Archives
Newsletters
Current Magazine Issue
Magazine Archives
Online Columnists
Ask the Experts
Industry News
Search DM Review

General Resources
Bookstore
Industry Events Calendar
Vendor Listings
White Paper Library
Glossary
Software Demo Lab
Monthly Product Guides
Buyer's Guide

General Resources
About Us
Press Releases
Awards
Media Kit
Reprints
Magazine Subscriptions
Editorial Calendar
Contact Us
Customer Service

Meta Data Solutions and Their Return on Investment

  Article published in DM Direct Newsletter
January 11, 2002 Issue
 
  By Adrienne Tannenbaum

The meta data solution is always a misunderstood benefit in most large organizations. Typically, the need for and value of meta data is equally as misunderstood; hence, the need or request for a cost justification of any expense that contains the word meta data.

In order to measure the return on investment of any meta data solution undertaking, it is essential that the depth and range of meta data solution coverage be clarified from the onset. Most meta data solution undertakings fall into one of the following three types (see Figure 1):


Figure 1: Types of Meta Data Solutions

Type 1: First-Time Meta Data Stores

Originally called data dictionaries, now expanded to cover meta data beyond the definitions and names of physically stored data elements. These meta data solutions are typically initiated by central data management, data administration functions. Originally, they were only common in very large organizations. Today, they are beginning to exist in any organization that can't find or interpret existing data. Because they take on an enterprise objective of sorts, they are the most likely to require cost justification, unless of course the mandate to initiate one comes from above. The distinguishing characteristics of this type of meta data solution are its intent and placement within the organization. Established as a means of standardization in most cases, they strive for the use of standard names, definitions and characteristics, all of which are tracked within this usually centralized meta data solution. Its creators have control and are typically the only ones permitted to establish and maintain these standards for read access throughout the organization.

Type 2: Data Warehouse Support

Data warehouses have resulted from the common inability to find, access and interpret corporate data in a timely manner. Based upon these common prerequisite characteristics, those organizations with foresight insist upon the delivery of meta data as part of the data warehouse architecture. Those organizations without foresight usually discover the need for meta data once the data warehouse is moved into production.

Organizations with unsuccessful or nonexistent Type 1 meta data solutions often create another meta data solution focused exclusively on meta data supporting the data warehouse only. In many cases, the meta data becomes tracked within a purchased data warehouse reporting, ETL or decision support product.

Type 2 meta data solutions are often the easiest to implement in that they typically require the least amount of overhead and additional cost. In most data warehouse efforts, the purchase of ETL or reporting software is a given, and vendors all offer built-in areas for meta data storage, which once populated are usually as accessible as the data to be analyzed. The scope rarely expands beyond warehouse coverage and explanation of the data that resides within the warehouse itself since, even when planned, the data warehouse reporting software is usually not capable of supporting a wide range of organization meta data.

Type 3: The Meta Data Solution

Very recent in intent, this meta data solution category is usually undertaken by those who have already undertaken initiatives of the first two categories and still are not experiencing meta data success. Once an organization's meta data becomes as disbursed and conflicted as its data, it is usually time to consider a global or connecting perspective. There are various types of meta data solution architectures and configurations, but they all share one common objective - that of keeping meta data where it is used and only standardizing "common meta data." Common meta data is almost always a small subset of the various meta data renditions that exist throughout a large organization.

Those organizations that embark on a Type 3 meta data solution are crying for help in more ways than one. The solution's sponsors, often the same suborganizations that created the Type 1 meta data solutions see a Type 3 solution as their saving grace. However, because the meta data situation in these organizations was usually never what it could have been, the initiating sub-organization is usually asked to justify this approach from both a time and a cost perspective.

Measuring the return on an organization's meta data solution investment, otherwise known as the ROI, requires the calculation of the specific investment cost, typically quite tangible, as well as the projected savings or increased earnings, often quite intangible. The investment is then evaluated against projected savings/earnings.

The Investment

What does it cost to build a meta data solution? Although we typically zero right in on the costs of purchased software and any associated consulting fees, many of us realize that a substantial portion of the initial costs occur (or should occur) before a software solution is even selected. Meta data requirements must be derived, and specific methodologies exist for obtaining the details related to the stated objective and the types of beneficiaries (people, tools, applications) that will be served by the to-be-implemented solution. The gathered meta data requirements are then sourced, categorized and organized and a series of meta models result. Other requirements (displays, access, security, interfaces, etc.) are then gathered, with a proposed architecture resulting. All of these efforts take time and involve resources, internal and often external.

Most organizations assign loaded dollar rates to internal employee time. External time (that of consultants and vendors) is usually based upon proposals and/or contracts. Very simply, the time that is required to perform the preliminary requirements analysis is multiplied by hourly (or daily) rates for each resource, with the result being equivalent to the labor cost. Other costs which may be involved include software and hardware purchases, post-installation vendor product customization/consulting fees, licensing costs, training and annual maintenance fees. Finally, if new employees are to be hired as administrators or meta data solution managers, these salaries clearly become an additional investment.

The resulting total should be evaluated by a standard time unit (e.g., cost per year) depending upon the amount of time required to achieve the initial meta data solution scope and subsequently extend or expand it. Generally, the first rollout should occur in no more than six months and should be scoped appropriately.

The Return

What triggered the beginnings of your meta data solution implementation? Whatever the major reason, it was costing the organization money. How much money? The dollar amount is often intangible in a sense, but always relates to inefficient processes (which require more time from more people) or incorrect decisions (based on the inability to locate or interpret suitable supporting data). In many situations these inefficiencies result in lost opportunities. Every information seeking or interpretation process which can be improved or streamlined by accurate and accessible meta data must be assigned a dollar cost.

Avoidable Costs

Assigning costs to information seeking and interpretation processes involves the collection of measurable as well as immeasurable dollar facts. Labor cost, as defined within the investment space is equally calculated as a projected savings. Costs, often targeted as potential savings, must be calculated for:

  • Locating Information - All time spent identifying potential sources of information, accessing (or attempting to access) these sources as well as acquiring external data has a cost. Resources, time spent per resource and the cost of any purchased software and external data used to help locate the targeted items all add up. These costs exist before the rollout of a well-implemented meta data solution and must be calculated as "projected savings."
  • Interpreting Information - Once information is found, it is often misinterpreted or not interpreted at all because its true meaning is not derivable. The phone calls, meetings, questions and required further analysis all require resources and time. In some situations, so much meta data exists that it is virtually impossible to figure out which instances to believe. More important, the results of misinterpretation are often extremely costly. In fact, one major product recall, manufacturing delay or missed opportunity that can be directly tied to the lack of accurate data is often the event that triggers the development of a meta data solution.
  • Integrating and Reorganizing Information - Putting data together for new analysis reasons requires the integration of multiple perspectives. Comparing perspectives and identifying their similarities and differences is a time-consuming task if accurate meta data is not as accessible as the source data itself. Even more frustrating is the fact that those involved in the integration efforts themselves often do not make their prospective decoding results available as meta data to those that access the integrated data result.

    Remember that the information and data that we seek and evaluate is not restricted to the numbers that we see on a report. Searches through any type of corporate information, including program libraries (using Y2K assessments as a perfect example) can benefit from organized meta data-based access. Again, depending upon the scope of the to-be-implemented meta data solution, the information of interest will form the boundaries for the projected savings. The scoped information's location and ease of access will also determine the amount of effort required.

    To put a dollar sign on the benefits of a meta data solution, assign labor costs to all of the above activities, as they are being performed within the organization. The number of resources, amount of time per resource and their actual loaded labor rates form the calculation as before. Also, any software purchases which were made based upon the lack of identifiable, accurate, accessible meta data (e.g., Y2K solutions) can also be earmarked as costs which could have been avoided.

    The formation of "information reporting" or "information research" groups often fits into another avoidable cost category when set up based upon the average manager's inability to find particular sets of information. Again, the number of employees assigned to this group along with their loaded salary rates forms another input into the overall avoidable cost column.

    Finally, the amount of time and resources that are assigned to product recalls, design reworks or application enhancements are often avoidable in that errors or oversights are typically based upon the lack of accurate supporting data. The costs of these mishaps are usually what trigger an effort of this kind to begin with.

Lost Revenue

How do we put a dollar amount on revenue that was missed due to misinterpretations or the lack of accurate and available decision making data? In some cases, these dollar amounts can be surmised based upon competitor sales. In other cases, industry forerunners have nothing upon which to base these estimates. Generally, the lost revenue number is an educated estimate, based upon experiences in the past and their impacts on the organization's income stream. Estimates should be focused however on all decisions that are or have been made with that information which needed the meta data targeted for coverage by the proposed meta data solution.

The ROI

Costs, benefits and break-even point all contribute to the estimated Return on Investment (ROI). Figure 2 illustrates the investment factors that contribute to the ROI calculation.

Tasks/Expense Type

Assigned Resource

Daily Loaded Salary

Extended Days

Total $

Requirements Analysis

Systems Analyst

$1,000

20

$20,000

 

Meta Data Specialist

$1,100

20

$22,000

 

Data Analyst

$750

35

$26250

   

Total Labor Costs

 

$68,250

         

Software

     

$85,000

Vendor Services

     

$15,000

First Yr. Maintenance

     

$10,000

   

TI Software Costs

 

$110,000

   

Total Investment

 

$178,250

Figure 2: Table 1 - Sample Investment Calculation

In Figure 2, an investment of approximately $175,000 will be required for the initially scoped meta data solution implementation. This investment does not include follow-on analysis and expansion, but does represent the start-up costs and initial annual maintenance fee. In addition, no permanent new hires were required in this example.

Calculating the projected cost avoidance within the scope illustrated above is depicted in Table 2 seen in Figure 3.

Information

       

Search/Locate

Systems Analyst

$1,000

10

$10,000

 

Marketing Analyst

$950

10

$9,500

 

Data Analyst

$750

15

$11,200

 

IT Developer

$650

5

$3,250

         

Information Analysis

Exec Mgmt.

$2,000

3

$6,000

 

Marketing Analyst

$950

3

$2,850

 

Data Analyst

$750

3

$2,250

 

IT Developer

$650

10

$6,500

         

Data Scrubbing

Data Analyst

$750

10

$7,500

 

Data Analyst

$750

6

$4,500

 

IT Developer

$800

10

$8,000

         

Data Redesign/Integrate

Systems Analyst

$1,000

10

$10,000

 

Data Analyst

$750

6

$4,500

 

IT Developer

$800

10

$8,000

 

IT Developer

$650

10

$6,500

   

Total Labor Costs

 

$100,600

         

Extraction Software

     

$45,000

Vendor Services

     

$15,000

   

TI Software Costs

 

$60,000

   

Total Cost Avoidance

 

$160,600

Lost Earnings

Late Product Entry

   

$2,500,000

 

Lost customers

   

$7,000,000

   

Total Lost Earnings

 

$9,500,000

   

Targeted $$ Return

 

$9,660,600

Figure 3: Table 2 - Sample Return Calculation

Figure 3 illustrates those additional tasks and expenses that typically result from the lack of accurate available meta data. The additional time marked for the "data redesign/integrate task" is reflective of time above and beyond the task's duration with easily locatable and understandable supporting data. As mentioned earlier, additional cost avoidances are often predicted in terms of lost revenue due to the fact that accurate data was not available to support these major business decisions.

Calculating the ROI is a simple division. The return is divided by the investment. In this example, there is no doubt that the ROI supports the meta data solution decision. However, support for such a concept is likely to be based upon an organization's experience with lost revenue as noted. Without such experiences, ROI will be based upon organizational efficiency (or lack thereof), and in many cases the ability to eliminate tasks and/or streamline staffing. Those organizations which have used organizational inefficiencies as a successful argument often have large reporting or data quality organizations charged with the tracking, validation and distribution of corporate data.

Conclusion

Calculating the ROI for a meta data solution effort springs from an organization's inability to use information for its intended purpose - the support of the business, its operations and its strategic growth. Putting dollars on these aspects of the business world is based upon cost avoidance, lost revenue and potential earnings. Dollars can be assigned to all of these using standard accounting principles.

...............................................................................

For more information on related topics visit the following related portals...
Meta Data and Strategic Intelligence.

Adrienne Tannenbaum is president of Database Design Solutions, Inc. (www.dbdsolutions.com), a New Jersey-based consulting firm specializing in the revitalization of corporate data. The firm focuses on data issues within large organizations and supports all data reconstruction efforts with a solid meta data backbone. Tannenbaum is the author of two popular meta data-focused books: Metadata Solutions: Using Metamodels, Repositories, XML, and Enterprise Portals to Generate Information on Demand (2001, Addison Wesley) and Implementing a Corporate Repository (1994, Wiley).

Solutions Marketplace
Provided by IndustryBrains

SAP Safe Passage for Current Customers
If your current applications are at risk, SAP Safe Passage provides a clear roadmap for solution migration with maintenance support & integration technology. View free demos now!

Design Databases with ER/Studio ? Download Now!
ER/Studio delivers next-generation data modeling. Multiple, distinct physical models based on a single logical model give you the tools you need to manage complex database environments and critical metadata in an intuitive user interface.

Save on Business Intelligence and Data Warehousing
Leverage Open Source database software and PC-based commodity hardware for an unsurpassed price/performance value. ExtenDB transforms the economics in developing a Business Intelligence infrastructure.

OrindaBuild Java source code generator for Oracle
OrindaBuild examines your Oracle database and creates a matching Data Access Object factory class and Web Service classes for your chosen SQL statements and PL/SQL procedures. Fully functional demo version available.

Help Desk Software Co-Winners HelpSTAR and Remedy
Help Desk Technology's HelpSTAR and BMC Remedy have been declared co-winners in Windows IT Pro Readers' Choice Awards for 2004. Discover proven help desk best practices right out of the box.

Click here to advertise in this space


E-mail This Article E-Mail This Article
Printer Friendly Version Printer-Friendly Version
Related Content Related Content
Request Reprints Request Reprints
advertisement
Site Map Terms of Use Privacy Policy
SourceMedia (c) 2005 DM Review and SourceMedia, Inc. All rights reserved.
Use, duplication, or sale of this service, or data contained herein, is strictly prohibited.