from information crisis to business excellence through...

P a r t

IFrom Information Crisis to

Business Excellence through Information Excellence

Chapter 1: “Process and Business Failure: the High Costs of Low Quality Information.” This chapter describes the omnipresence of Poor Quality Information and the process and business failure it causes, economically to the enterprise and socially as harmful to end Customers, Investors, Citizens, and Taxpayers.

Chapter 2: “The ABCs of the TIQM Quality System for Total Information Quality Management.” This chapter describes the proven principles, processes, and techniques of sound Quality Systems, such as Deming, Juran, Shewhart, Ishikawa, Crosby, Imai (Kaizen), Taguchi (Quality Engineering), Akao (Quality Function Deployment), The Baldrige Criteria, Six Sigma, and others, which must be applied to Information Process to address our Information Quality (IQ) problems and the Process and Business failures they cause. We describe these IQ Principles and Processes in Part I.

34474c01.indd 1 9/1/09 8:39:46 AM

COPYRIG

HTED M

ATERIAL

34474c01.indd 2 9/1/09 8:39:46 AM

3

At about the time I started writing this book, I received an email from an Information Professional, G.P. Here are the edited words of G.P.:

“My career in Information Management has now taken me from [Company 1], to [Company 2], to [Company 3], to [Company 4]. The thing that they all have in common is a desire to cut corners and deal with quality later. It takes a lot of energy to be the Information Quality cheerleader, and I find it discouraging and overwhelming at times. Keep writing your articles and books to encourage all the people like me who are dealing with these issues every day.”

G.P., this book is written especially for people like you who care about your end-Customers and your internal Information Consumers, who require Quality Information to perform their work effectively to accomplish the enterprise mis-sion and to stay in business.

The principle of management by quantity or speed over quality is a failed management philosophy.

What is our experience of the costs of poor quality information?In this chapter I provide specific real-world anecdotes of the high costs of low

quality Information and Information Systems.

C h a P t e r

1Process and Business Failure:

The High Costs of Low Quality Information“In the U.S.A., about a third of what we do consists

of redoing work previously ‘done’.” 1

Joseph M. Juran

34474c01.indd 3 9/1/09 8:39:46 AM

4 Part I n Information Crisis to Business Excellence

In virtually every disaster, Information Systems and Information Quality failures alike, there will be losses of Customers due to loss of confidence in the companies. British Gas, for one example, lost over 1 million Customers in 2006 alone. The lost revenue from this will exceed the direct cost losses from the failure.

Table 1-1 lists anecdotes of poor quality Information System Software and the costs of poor quality. The source and dates of the anecdotes are found at the end of each Outcome column cell entry. See the legend at the end of the Table.

Table 1-1: The High Costs of Poor Quality Information Systems Software and Databases

ComPany / IndusTry

ouTComE (CosTs (m = mIllIon; B = BIllIon; T = TrIllIon) (rEf/daTE))

us$ loss amT

Allstate Insurance Co. Office Automation System abandoned after deployment, costing $130 M. (WSF, 1993).

$130,000,000

Arianespace [France] Software specification and design errors cause $350 M Ariane 5 rocket to explode. (WSF, 1996).

$350,000,000

AT&T Wireless Customer Relationship Management (CRM) upgrade problems lead to rev-enue loss of $100 M. (WSF, 2003–4).

$100,000,000

Avis Europe PLC [UK] Enterprise Resource Planning (ERP) System canceled after $54.5 M spent. (WSF, 2004).

$54,500,000

High Costs and Losses of Poor Quality Information

Here are sample anecdotes collected over the past twenty years that illustrate the costs and consequences of poor quality information and Information Systems.

Table 1-1 highlights the high costs of poor quality in Information Systems.

dIsClaImEr

The following anecdotes in Tables 1-1 and 1-2 have been cited from reputable sources. While I have not audited these findings, I have no reason to doubt their accuracy to within a 90 to 95 percent confidence level. I have sought to calculate the exchange rates at the prevailing time period of the failures.

Some of the examples listed are associated with biased and misleading information or fraud. Fraud, of course, is the worst scenario, and some may challenge them as not applicable. However, deliberate distortion of information also has costs to society, hurting individuals, organizations, and society as well. These truly appear as costs, financially, psychologically, and sociologically.

34474c01.indd 4 9/1/09 8:39:46 AM

Chapter 1 n Process and Business failure: The High Costs of low Quality 5

ComPany / IndusTry


us$ loss amT

British Gas (Centrica) £300 M overhaul of billing System over-bills Customers. BG wrote off £200 M due to complaints (11, 2006–2008) and filed suit against its contractor for £182M. (£482M * $2/£).

$964,000,000

BSkyB After a failed Customer Relationship Management system, BSkyB sued EDS for £709 M ($1.349 B($1.9)), for misrep-resented EDS’ abilities. (11, 2008/05/13).

$1,349,000,000

Budget Rent-A-Car, Hilton Hotels, Marriott Int’l, and AMR

American Airlines Travel Reservation System canceled after $165 M spent. (WSF, 1992).

$165,000,000

Chemical Bank Software error causes $15 million to be deducted from 100,000 Customer accounts. (WSF, 1994).

$15,000,000

CIGNA Corp. Problems with CRM System contribute to $445 M loss. (WSF, 2002).

$445,000,000

Federal Bureau of Investigation

$170 M for Networked System to replace paper system. Scrapped one year after delivery. (Washington Post, 2006/08/18).

$170,000,000

Ford Motor Company Purchasing System abandoned after deployment costing $400 M. (WSF, 2004).

$400,000,000

FoxMeyer Drug Company

$40 M ERP System abandoned after deployment, bankrupting company. (WSF, 1996).

$40,000,000

Greyhound Lines Inc. Bus Reservation System crashes repeat-edly upon introduction, causing revenue loss of $61 M. (WSF, 1993).

$61,000,000

Hershey Foods Corp. Problems with ERP System contributes to $151 M loss. (WSF, 1999).

$151,000,000

Hewlett-Packard Company

Problems with ERP System contribute to $160 M loss. (WSF, 2004).

$160,000,000

Hudson Bay Co. [Canada]

Problems with Inventory System contrib-ute to $33.3 M loss. (WSF, 2005).

$33,300,000

J Sainsbury PLC [UK] Supply-Chain System abandoned after deployment costing $527 M. (WSF, 2004).

$527,000,000

(continued)

34474c01.indd 5 9/1/09 8:39:46 AM


ComPany / IndusTry


us$ loss amT

Kmart Corporation Supply-Chain System canceled after $130 M spent. (WSF, 2001).

$130,000,000

London Ambulance Service [UK]

Dispatch System canceled in 1990 at $11.25 million; second attempt aban-doned, costing $15 M. (WSF, 1993).

$26,250,000

London Stock Exchange [UK]

Taurus Stock Settlement System can-celed after $600 M spent. (WSF, 1993).

$600,000,000

McDonald’s Corporation

The Innovate Information-Purchasing System canceled costing $170 M. (WSF, 2002).

$170,000,000

Nike Inc. Problems with Supply-Chain System causes $100 M loss. (WSF, 2001).

$100,000,000

Oxford Health Plans Inc.

Billing and Claims System problems con-tribute to quarterly loss; stock plummets, leading to $3.4 B loss in corporate value. (WSF, 1997).

$3,400,000,000

Snap-on-Tools Problems with Order-Entry System causes $50 M revenue loss. (WSF, 1998).

$50,000,000

State of California DMV System canceled after $44 M spent. (WSF, 1994).

$4,000,000

State of Mississippi Tax System canceled after $11.2 M is spent; state receives $185 M in damages. (WSF, 1999).

$196,200,000

State of Washington Department of Motor Vehicle (DMV) System canceled after $40 M spent. (WSF, 1997).

$40,000,000

Sydney Water Corp. [Australia]

Billing System canceled after $33.2 M spent. (WSF, 2002)

$33,200,000

Toronto Stock Exchange [CAN]

Electronic Trading System canceled after $25.5 M is spent. (WSF, 1995).

$25,500,000

UK Inland Revenue S/W errors contribute to $3.45 B tax-credit overpayment. (WSF, 2004–5).

$3,450,000,000

U.S. Federal Aviation Admin.

Advanced Automation system canceled after $2.6 B spent. (WSF, 1994).

$2,600,000,000

U.S. Internal Revenue Service

Tax Modernization effort canceled after $4 B is spent. (WSF, 1997).

$4,000,000,000

Table 1-1 (continued)

34474c01.indd 6 9/1/09 8:39:46 AM


ComPany / IndusTry


us$ loss amT

United Way Administrative Processing System can-celed $12 M spent. (WSF, 1999).

$12,000,000

Washington, D.C. City Payroll System abandoned after deployment $25 M spent. (WSF, 2000).

$25,000,000

U.S. Bureau of Indian Affairs

Spent $12.5 M on Maintenance Repair System for schools that failed to capture quality information. (5: 2003/08).

$12,500,000

U.S. Business spend on unwanted systems

BPN Forum survey = 10% to more than 20% of IT budget wasted on unwanted systems; Calc: $13.84 T (U.S. GDP) × 10% (IT budget) = $138.40 B X 20% (waste rate/yr) = $24.6 B wasted per year. (BPN Forum, ).

$24,600,000,000

Total Costs of Poor Quality Software in these examples in US$

$44,589,450,000

Sources: WSF: “Why Software Fails”; BW = Business Week; CEO = CEO Magazine; CW = Computerworld; IW = InfoWeek; FCW = Federal Computer Week; For = Fortune; NYT = The New York Times; Tim = Time; WSJ = The Wall Street Journal; III = Information Impact International; USA = USA Today; FT = Financial Times; LTim = London Times; LTel = London Telegraph;TN = The Tennessean; N&O = News and Observer; Aus = The Australian; AFR = Australian Financial Review; LAT = LA Times; NAO = UK Nat’l Audit Office; Wired.com = http://www.wired.com/science/discoveries/news/2004/02/62242.

Table 1-1 illustrates the incredibly high Costs of Poor Quality in Information Systems software. These are Information Systems with such poor quality that they had to be scrapped. For every scrapped Information System, there are hun-dreds more that have significant defects that Knowledge Workers have to live with. Information Producers and Knowledge Workers are struggling to perform their work when the defects are not so catastrophic as to scrap the systems. This wastes the time of our most important resource: people.

As organizations become more and more dependent on Information Technology, they must ensure they design quality into the Information Management and Information Systems Engineering processes, so they error-proof Information Systems to the extent possible.

When systems fail, business processes stop or must be handled manually, always with a risk of omission and error in capturing data.

Table 1-2 documents anecdotes of actual losses and Costs of Poor Quality Information that cause business processes to fail, requiring costs of recovery from the failure and the Information Scrap and Rework required to find and correct the information so other processes do not fail. Sources and dates of anecdotes are found at the end of each Outcome cell in Table 1-2.

34474c01.indd 7 9/1/09 8:39:46 AM


Table 1-2: The High Costs of Poor Quality Business Information

ComPany / IndusTry

ouTComE (CosTs (rEf daTE)) loss amounT

AIG Agrees to pay $1.6 B to settle allega-tions that it used deceptive accounting practices to mislead investors and reg-ulatory agencies. (TN, 2006/02/10).

$1,600,000,000

AIG Bailout money of $223 B from taxpay-ers. (TN, 2009/03/03).

$223,000,000,000

Arthur Andersen Auditing failure at Enron led to $1.3 B loss in pension. (FT, 2002/01).

$1,300,000,000

AT&T AT&T agrees to pay $177,000 in fines in Tennessee for billing Customers for services they hadn’t authorized. (TN, 2004/03/06).

$177,000

Australia Dept. of Defence

Defence to lose compensation billions in reforms due to overcompensation due to error in inflation: A$ 4.3 B = US$ 2.99 B. (AFR, 2009/03/30).

$2,990,000,000

Bristol-Myers-Squibb Fined $150 M for inflating its revenue. (USA, 2004/04/05).

$150,000,000

Bristol-Myers-Squibb Paid $300 M to settle lawsuit (accounting re: Imclone). (USA, 2004/04/04).

$300,000,000

Cell Phone Companies

Agree to pay 32 states $5 M and required to give Customers a grace period for cancellations and better information. (TN, 2004/07/22).

$5,000,000

City Government Free garbage pick-up due to oversight for 90 homes outside of Nashville’s urban services district for almost 10 years may have cost the city as much as $55,000. (TN, 2005/03/10).

$55,000

Citigroup Citigroup paid $2 B to settle a lawsuit over its role in helping Enron manipu-late its accounting that led it to Enron’s collapse. (TN, 2005/06/11).

$2,000,000,000

Citigroup Citigroup paid $2.7 B to investors to settle a class-action lawsuit over losses in WorldCom accounting fraud. It set aside another $6.7 B for other lawsuits. (USA, 2004/05/11).

$9,400,000,000

34474c01.indd 8 9/1/09 8:39:47 AM


ComPany / IndusTry


Citigroup Paid $18 M refunds and settlement for taking $14 M from Customers’ credit card accts. (USA, 2008/08/27).

$18,000.000

Community Health Systems

CHS to pay $31 M in settlement due to billing errors. (TN, 2000/03).

$31,000,000

DirecTV Paid $5 M for misleading program packaging. (TN, 2005/12/03).

$5,000,000

DuPont DuPont’s failure to report widespread exposures to a toxic chemical costs it $16.5 M in fines & payouts. (TN, 2005/12/13).

$16,500,000

Enron Ex-CEO Skilling enters prison for 24-year sentence. He and Ken Lay were convicted on many counts of fraud, conspiracy, and insider trading — loss of $60 B in stock & $2 B in pen-sion plans. (TN, 2006/12/14).

$62,000,000,000

Enron Settled lawsuit for $168 M for mislead-ing Stakeholders. (TN 2005/01/09).

$168,000,000

Fannie Mae Required to restate 2004 earnings down by $11 B due to serious account-ing problems and earnings manipula-tion. (TN 2006/05/10).

$11,000,000,000

Fannie Mae Spent $1 B to correct years of false financial reports. (TN 2008/08/24) .

$1,000,000,000

Federal Agency Overpaid one set of beneficiaries by $1.7 B and underpaid another set of beneficiaries by $600 M. (Private Client, 2005).

$2,300,000,000

FEMA The GAO 95% confident that improper FEMA payments were between $600M & $1.4B re Katrina & Rita = $1 B. (USA, 2007/06/15).

$1,000,000,000

Financial A Acquired subprime mortgage company without due diligence on data. Write-down $2.6 B & $1 B in debt & 100 M loss. (III, BW, 2000).

$3,700,000,000

Franklin Advisors Settled lawsuit for $50 M for illegal, rapid trading transactions. (USA, 2004/08/03).

$50,000,000

(continued)

34474c01.indd 9 9/1/09 8:39:47 AM


ComPany / IndusTry


General Motors Takes a $39 B charge to stay in compli-ance with accounting law. (USA, 2007/11/07) .

$39,000,000,000

Goodyear Billing errors in accounting led to $100 M loss. (TN, 2003/10/23).

$100,000,000

Google Shortchanged thousands of their advertisers and settled a lawsuit for $60M. (2006/05/08).

$60,000,000

Government Billions in 9/11 recovery loans were misused by those who were not affected or did not need them. Estimated at $2.5 B. (TN, 2005/09/09).

$2,500,000,000

Government Contractors

U.S. Military suspends $160 M in charges due to inaccurate and incom-plete files and bills. (USA, 2004/05/18).

$160,000,000

Halliburton Report questions over $1.4B of Halliburton Bills. (USA, 2005/06/28).

$1,400,000,000

Halliburton KBR Poor accounting costs KBR $1.2 B for contracting. (WSJ-EU, 2004/04).

$1,200,000,000

Healthways Calls off $307.5 M merger with LifeMasters Supported Self Care over “data & reporting errors” that mini-mized its value. (TN, 2006/08/03).

$307,500,000

Hubble Telescope Error in mirror specification causes a $700 M repair. (wired.com, 2004/02/11).

$700,000,000

Institute of Medicine Report found 1.5 million people harmed by medication errors per year. Extra costs to treat medication-related inju-ries in hospitals alone amount to at least 3.5 B per year. (USA, 2006/07/21).

$3,500,000,000

Insurance company A Lost $5 B one year and $3 B the next due to poor risk analysis. (III Private, 2002).

$8,000,000,000

Insurance company B One third of staff time is Information Scrap and Rework = $23.4 M/year. (III Private, 2004).

$23,430,000


34474c01.indd 10 9/1/09 8:39:47 AM


ComPany / IndusTry


Janus Capital Client pulls $5 B out of Janus over improper trading. Losses est. at $250 M/yr. (TN, 2004/07/30).

$250,000,000

Janus Capital Group Fined $226 M for improper trading actions. (WSJ, 2004/04/20).

$226,000,000

KPMG Sued $1 B (£692 M) for accounting irregularities in collapse of New Century Financial. (FT, 2009/04/02)

$1,000,000,000

L.A. County Due to poor calculation of pension funds incurred $1.2 B liability. (LAT, 1998/04/08).

$1,200,000,000

Lucent Fined $25 M for overstating revenue and earnings. (USA, 2004/05/18).

$25,000,000

Mazda Fined $5.25 M for failure to disclose required information on leases. (TN, 1999/10/02).

$5,250,000

Medical Care 98,000 people a year in U.S. die from medical error @ $147 M legal costs. (USA, 2002/07/24).

$147,000,000

Medicare 35M Handbooks with wrong info. 13,000,000 Booklets @ $.55/book to replace. (TN, 2005/10/6).

$19,250,000

Medicare Paid out $339 M in improper claims. (TN, 1998/10/01).

$339,000,000

Merck Merck paid IRS $2.04 B due to disal-lowed deductions. (WSJ, 2004/05/10).

$2,040,000,000

Merck Merck & Co. settled its U.S. personal injury claims over Vioxx for $5.2 B in 11/2007 (deceptive advertising). (Aus, 2009/03/30).

$5,200,000,000

Merck Merck & Co. seeks to settle its personal injury claims over Vioxx in Australia. Estimated legal costs @ US $ 400 M. (Aus, 2009/03/30).

$400,000,000

Mizuho Securities Typing error leads to 225M loss on Tokyo stock exchange. Trader tried to sell 610,000 shares at 1 yen of com co. He intended to sell 1 share at 610,000 yen. (TN, 2005/12/11).

$225,000,000

(continued)

34474c01.indd 11 9/1/09 8:39:47 AM


ComPany / IndusTry


Morgan Stanley Paid $15M fine to settle charges that it hid a large trove of archived informa-tion from the U.S. Securities and Exchange Commission. (USA, 2006/05/11).

$15,000,000

Morgan Stanley Books a quarterly loss of at least $1.2 B to $1.7 B due to wrong accounting treat-ment of bonds. (WSJ.com, 2009/04/09).

$1,450,000,000

Morgan Stanley Fined $2.2 M for late communication to investors. (WSJ, 2004/07/30).

$2,200,000

Mutual Fund Companies

Fleeing investors hurt scandal-tainted fund companies worse than fines. Losses estimated at $500 M. (TN, 2004/09/03).

$500,000,000

NASA-Mars Climate Orbiter

Miscalculation of English units of mea-sure to metric causes total loss of $125 M and lost scientific measure-ment. (CNN.com, 1999/09/30).

$125,000,000

Nashville Metro Govt. Overlooked collection of fines and has lost $57 M. (TN, 2009/01/16).

$57,000,000

Natwest Bank Derivatives pricing error led to £91 M = $152 M loss; selling off Equities divi-sion led to $1.192 B write-off (Tot = $1.344 B). (USA, 1997/12/03).

$1,344,000,000

Not-for-Profit org. Direct Costs due to poor IQ = $42,192,000 per year. (III Private, 2008).

$42,192,000

Package Delivery co. Failed to invoice for 4.2% of packages delivered = $340 M/year. (III Private, 2006/02).

$340,000,000

Pentagon Pentagon’s books defy auditors; fixing it costs at least $13 B. (N&O, 2006/02/12).

$13,000,000,000

Pharmaceuticals fined

Abbott fined $600 M for improper marketing of some products. (USA, 2003/07/24):

AstraZeneca fined $354 M for giving drugs to MDs to bill medicare;

Ross Products paid $200 M and $400 M for criminal/civil penalties;

Tap Pharmaceuticals paid $875 M fine.

$2,429,000,000


34474c01.indd 12 9/1/09 8:39:47 AM


ComPany / IndusTry


Putnam Investments Customers withdrew $4.4B from Putnam funds over accounting errors. (WSJ, 2003/11/7).

$4,400,000,000

RiteAid Errors in accounting led to $1.81 B write-down over three years. (TN, 2002/6/22).

$1,810,000,000

Shell Fined $151.5 M for overstating oil reserves. (WSJ, 2004/07/30).

$151,500,000

Sprint Sprint records $29.5 B loss in Nextel write-down. Poor service and dropped calls lost Sprint 683,000 subscribers. (Blm, 2008/02/29).

$29,500,000,000

State of Tennessee Had to compensate wrongly convicted man $833,000. (TN, 2004/10/05).

$833,000

Telco A Costs of poor quality eliminated over 10 years $800 M. (III Private, 2004).

$800,000,000

Telco B Losses of $50 M in un-invoiced ser-vices (III Private, 2002).

$50,000,000

Ten Wall Street firms Fined a total of $1.4 B by SEC (mis-leading investors) (TN, 2004/08/27).

$1,400,000,000

UBS Lost $664 M (911 M Swiss francs) due to poor risk management data. (NYT, 1998/11/18).

$664,000,000

UBS Lost 19.7 B Swiss francs ($17 B) in 2008 as it wrote down the value of some debt assets and wealth manage-ment clients withdrew money. (NYT, 2009/5/29).

$17,000,000,000

U.K. Gov. Pension calculation errors cause some pensioners to be overpaid by up to £140 M with others underpaid by £100 M. (LTim, 2009/3/9). Tot £240 M x 1.45 (exchange) = $348 M.

$348,000,000

U.K. Inland Revenue Inland Revenue overpaid tax credits for disabled persons of between £510 million and £710 million from errors in understated income.£82M ($123.3 M). (NAO, 2003/11/19). (exchange= $1.45:£= $884.5 M).

$884,500,000

(continued)

34474c01.indd 13 9/1/09 8:39:47 AM


ComPany / IndusTry


U.K. Inland Revenue Penalties caused by failed Inland Revenue processing. 300,000 at £900 ($1.504) = $406 M. (LTel, 2003/03).

$406,080,000

U.K. Office of Nat’l Statistics

Overpaid staff £11 M ($16.5 M). (FT, 2003/03/26).

$16,544,000

U.K. Revenue & Customs

A National Audit found £6.6 B overpay-ment in tax credits; £1-1.3 B paid in 2004–5 to claimants not entitled; £575 M in 2006; £800m not collected and £340 M overpaid by taxpayers. £135 M (US$ 191 M) a year lost because tax due on pensions is not collected. (FT, 2007/7/13). £1 = $2.037. (exch on 2007/07).

$17,926,000,000

U.S. Bureau of Indian Affairs

After spending $12.5 M on the system software, they had to spend $13 M to correct the records. (FCW, 2003/08).

$13,000,000

U.S. Government Costs to upgrade voting equipment: $3.8 B after the 2000 Presidential election failure. This did not solve the problem. (USA, 2002/10).

$3,800,000,000

U.S. Government Owed more than $35 B in fines from criminal and civil cases but collects less than half of that ($20 B). (TN, 2006/03/09).

$20,000,000,000

U.S. Government An error in an addendum that required oil and gas companies to pay federal royalties on offshore leases cost the US $7B (TN, 2006/03/02).

$7,000,000,000


A study conducted by Treasury Department investigators posing as taxpayers determined that IRS employees gave wrong answers to 28% of the questions. $50 M est. losses. (TN, 2003/09/04).

$50,000,000


The U.S. tax gap of taxes owed to taxes collected is a deficit of $290 B per year in 2001 and 2002. (TN, 4/17/2006).

$290,000,000,000


34474c01.indd 14 9/1/09 8:39:47 AM


ComPany / IndusTry



“Tax gap” for the IRS in failure to collect taxes is around $350 B per year in lost revenue in 2005 and 2006. (S. Hrg. 110-698, 2007/04/18). (http://finance.senate.gov/hearings/46363.pdf ).

$350,000,000,000


Loses from $200 to $300 M in taxes because software that screened for fraudulent refunds was not working. (TN, 2006/07/15).

$250,000,000

University of California

Sued AOL for overstating ad revenues and subscribers. Loss of $450 M in value. (USA, 2003/04).

$450,000,000

WorldCom Former WorldCom head, Bernie Ebbers jailed over the $11 B account-ing manipulation. (FT, 2005/03/16).

$11,000,000,000

WorldCom Fined $500 M by SEC for its $9 B accounting scandal. (USA, 2003/05).

$500,000,000

Xerox and Arthur Andersen

Xerox fined $10 M and Andersen fined $7 M for inaccurate financial reporting. (USA, 2002/04).

$17,000,000

subtotal Poor Quality Information Costs — from Table 1-2

$1,167,785,029,000

Poor software Quality Costs

Poor Quality Software Costs — from Table 1-1

$44,589,450,000

Grand Total Total Costs of Poor Quality Information & Systems

$1,212,374,479,000

Sources: WSF: “Why Software Fails”; BW = Business Week; CEO = CEO Magazine; CW = Computerworld; IW = InfoWeek; FCW = Federal Computer Week; For = Fortune; NYT = The New York Times; Tim = Time; WSJ = The Wall Street Journal; III = Information Impact International; USA = USA Today; FT = Financial Times; LTim = London Times; LTel = London Telegraph; TN = The Ten-nessean; N&O = News and Observer; Aus = The Australian; AFR = Australian Financial Review; LAT = LA Times; NAO = UK Nat’l Audit Office; S. Hrg. = Senate Hearing; Wired.com = http://www.wired.com/science/discoveries/news/2004/02/62242.

If the costs of Poor Quality Information of: $1,212,374,479,000 — nearly One and a Quarter Trillion Dollars — of waste in these 122 mostly reputable organizations, not counting recurring annual costs, doesn’t shock the heck out of you, I do not know what will motivate you to address Information Quality within your own organization.

34474c01.indd 15 9/1/09 8:39:47 AM


Many organizations may read about these companies and government enti-ties and write them off. “But we are different,” they may say. But if they are not in Stage 3 of Information Quality Management maturity, they will only be misleading themselves.

Remember that these are direct costs only and do not list missed or lost oppor-tunity. Also know that your costs to create and develop a sustainable Information Quality Management capability and culture will be a fraction of the waste you are incurring every year in out-of-control information processes.

Also know that you cannot solve your problems by simply buying IQ software and inspecting and correcting your way out of your IQ problems. You must attack this as a serious threat to your enterprise, with your IQ initiative being driven by your Top Leadership Team.

Societal Costs and Losses Caused by Poor Quality Information

It is not just the direct and opportunity costs of poor quality information to the enterprise that are harmful, it is also the societal losses, causing Customers to pay higher prices for goods and services and causing Citizens to pay higher taxes. Defective information can cause death and injury due to medication error or wrong medical procedures. Students may be denied entrance to universities because of errors in scoring required entrance exams. Consumers may buy harmful foods if nutritional labels are not accurate to a minimum level. People will be convicted of crimes they did not commit or guilty offenders may be released in error, putting citizens at risk. See the following sidebars for a small, but representative, examples of such losses.

sTudy: mEdICal Errors KIll as many as 98,000 u.s. PaTIEnTs yEarly

Journal of the American Medical Association reports as many as 98,000 people in the U.S. die from medical errors, despite increased focus on attempts to reduce errors.2

u.s. navy suB HITs mounTaIn noT on ITs navIGaTIon CHarTs

The nuclear attack submarine, the San Francisco, was severely damaged when it cruised at high speed, around 30 knots, into an underwater mountain that was not on its navigation charts. One sailor was killed, while 23 others were injured. Fortunately, the nuclear reactor was not damaged. If damaged, it could have caused long-term problems.3

34474c01.indd 16 9/1/09 8:39:47 AM


raPIsT rEmaInEd aT larGE for EIGHT yEars BECausE of “TyPInG Error”

A violent rapist in the U.K. remained at large for eight years because of a “typ-ing error” in recording DNA evidence, police said. The rapist attacked and tried to strangle a 19-year-old woman who was walking home in Rainham, Kent, in 1996. But the rapist was not arrested for some time, despite giving a DNA sam-ple that matched police evidence three months after the attack. He was jailed for 12 years by Maidstone Crown Court after admitting to the crime.5

faTal aEroBaTICs as navIGaTIon maP omITs Gondola loCaTIon

Maps given to Marine flight crews did not show a gondola lift on the steep mountainside at Cavalese in the Italian Alps, even though lifts have been there for decades. Also, the flight rules issued did not include a ban on flying below 2,000 feet above the terrain. The Marine pilot was showing off that day, flying faster than authorized and dipping below 400 feet above the terrain.

Whether he was aware of the gondola, his reckless behavior led to his slic-ing the gondola cable, causing the car to come crashing down, killing the 20 people on board.4

dIalysIs EQuIPmEnT sEnT To WronG PlaCE

Four new dialysis chairs were sent to a remote hospital in north Queensland, Australia, where no patients with kidney problems lived. At the same time, no dialysis chairs were provided in locations where many dialysis candidates reside.6

Euro CoIns ordErEd To BE rEmInTEd

Nine million of the first Euro coins minted in France had to be melted down and minted again in 1998. Some of the coins were not minted to specification. The 10-cent and 50-cent coins had to be reminted because they did not conform to the previously agreed specifications that called for raised ridges along the edges.

Vending machine operators found that the 50-cent coin was too light, mak-ing it impossible for machines to tell it apart from the 20-cent coin.

This example illustrates the importance of correct product specifications.7

smarT BomBs dEsTroy HousE, KIllInG InnoCEnTs

In January 2005, U.S. forces mistakenly dropped a 500-pound bomb on a house, destroying it and killing from five to 14 people, injuring five others. The house was not the intended target.

There have been many unfortunate failures of so-called “smart bombs” and attacks on innocents, due to poor quality information and communication in the Afghanistan and Iraq “wars.”8

In another incident, a U.S. pilot failed to clearly identify a target and fired on friendly forces, killing a British soldier and injuring several others.9

34474c01.indd 17 9/1/09 8:39:47 AM


PassEnGEr WEIGHT a CausE of CrasH of flIGHT us5481

US Airways Express flight 5481 came crashing down just after take off due to two causes. One was the inaccurate average weight of passengers was too low, and the other was that poor maintenance of the cables that attached to the elevator that adjusted the plane’s tail flap. As a result of these two prob-lems, the pilots were unable to control the flight, killing all 21 people aboard.

As the American population has grown, literally, in weight, airlines must measure their passengers’ weight and adjust the weight measurements to ensure safety of all. The FAA issued a temporary order to add 10 pounds to each passenger’s weight and five pounds to the weight of each bag.10

InCorrECTly GradEd saT TEsT sCorEs loWErEd sCorEs for 4000

Errors in software that graded the SAT exams of U.S. college applicants in October 2005 caused 4,000 exam takers’ scores to be wrong, lower than their actual results. In 5% of the tests, scores were 100 points lower than they should have been. This could have caused college applicants to be denied entrance.

In another incident in Minnesota, a scoring error led to 8,000 students being told they failed a 2000 state math test.

In 2005, 4,000 students taking an ACT Chemistry exam were given wrong scores.11

mIslEadInG food laBEls lEad To Poor HEalTH dECIsIons

Nutritional information on some food products is “wildly inaccurate,” a U.K. consumer magazine has warned. They looked at 570 nutrients in 70 products and found just 7% exactly matched the quantities on the labels. Some 17% fell outside the accepted 20% margin of error, including a kids’ pizza with 47% more sugar than stated.

Currently there is no specific law about how accurate the information on food labels should be. They need only show average nutrition values in the U.K.

Worst offenders:

Rivington’s Pink Panther wafers had three times more saturated fat ■n

than stated.

Cadbury’s Light Trifles contained 23% more fat than stated.■n

Tesco Kids Hot Dog Pizza contained 47% more sugar than claimed.■n

The consequences are that food choices are made based on misleading nutritional information, which will cause health problems.

Editor Malcolm Coles said: “Nutrition labels help people compare foods and make healthy choices, but only if they’re accurate. How can you trust what you’re eating when so many labels fall outside even the fairly generous mar-gins of error allowed?”12

34474c01.indd 18 9/1/09 8:39:48 AM


28% of moTor lICEnsEs In souTHWEsT EnGland arE noT aCCuraTE

A government study reveals that almost 28% of all drivers in the region do not update their licenses when they get married or move. Accurate information is required for police to track down stolen vehicles and for emergency crews at the scene of an accident. Failure to keep data accurate can lead to a £1,000 fine.13

fIrm sEEKs 200,000 TIrEs suBjECT To rECall

Poor quality information is hampering Bridgestone/Firestone from locating some 200,000 defective recalled tires that may still be on Ford Explorers. Owners may be at risk if they are driving on them, and are urged to notify Bridgestone/Firestone.14

loss of nasa’s mars ClImaTE orBITEr Is loss of KnoWlEdGE

While the loss of the $125 million Mars Climate Orbiter wasted money, the real loss was the inability to capture knowledge about Mars to better under-stand our universe.15

The Orbiter’s loss was caused by the use of English units of measure to cal-culate the trajectory and altitude of the Orbiter as it was supposedly sent into orbit around Mars. However, the NASA team entered the data as metric data (NASA’s standard). By the time the NASA team realized that the data provided was not metric, it was too late to transmit the corrected data, and the Orbiter was sent in to an altitude below the gravitational pull, sending it crashing down on the backside of Mars.

a BroKEn sysTEm: Error raTEs In u.s. CaPITal CasEs 1973–1995

Overall rate of prejudicial error (serious, reversible error) in the ■n

American capital punishment system was 68%.16

Capital trials produce so many mistakes it takes three judicial reviews ■n

to catch them.

Of the 2,370 death sentences thrown out due to serious error, 90% ■n

were overturned by state judges; many had imposed the sentence in the first place.

Most common errors:

Defense lawyers didn’t look for, or missed evidence that the defendant ■n

was innocent.(continued)

34474c01.indd 19 9/1/09 8:39:48 AM


Police/prosecutors discovered missed or important evidence, but ■n

suppressed it.

82% of the people whose capital judgments were overturned by state ■n

post-convictions courts due to serious error were found to deserve a sentence less than death.

In 20 out of 23 study years, more than 50% of capital cases had serious ■n

flaws. In 10 of the study years, the error rate was 60%.

Over 90% of death-sentencing states have error rates of 52% or higher. ■n

85% have rates of 60% or higher. Three-fifths have error rates of 70% or higher.

BroKEn ElECTIons ProCEssEs dIsEnfranCHIsE mIllIons of voTErs

The 2000 Presidential election saw from four to six million voters lose their votes because of over-voting (voting for two or more candidates) or under-voting (not recording a vote, failed chad penetration, or too light pencil marks). The number of disenfranchised voters was the equivalent of more than all the voters who voted in Florida that year!17

25 PErCEnT of u.s. ConsumEr CrEdIT rEPorTs ConTaIn maTErIal Errors

Conducted by the U.S. PIRG Education Fund studied reports from thirty states and found 25 percent contained material errors that would deny credit for the individuals affected. In addition:

54% contained personal information that was inaccurate or missing.22 % of reports listed the same mortgage or loan twice.8 % were missing major credit, loan, or other consumer accounts that dem-

onstrated creditworthiness of the consumer.30 % contained credit accounts that had been closed by the consumer but

remained listed as open.Altogether, 79 % contained serious errors or other mistakes.18

The sad truth in all of these examples and many others ends up costing the Consumers, Citizens, innocent bystanders, and Voters in time and money, pay-ing for the recovery from process failure and Information Scrap and Rework because of organizations’ broken information processes.

a BroKEn sysTEm: Error raTEs In u.s. CaPITal CasEs (continued)

34474c01.indd 20 9/1/09 8:39:48 AM


Revisiting the Costs of Poor Quality Information

The costs of poor quality in manufacturing, service, and government have been well documented over the past seven decades. Table 1-3 lists what the quality gurus peg as the costs of poor quality.

Table 1-3: The High Costs of Low Quality in Manufacturing and Service Sectors

QualITy auTHorITIEs IndusTry

PErCEnT of rEvEnuE

loW To HIGH

Philip Crosby Manufacturing 15% to 20%

Siegel & Shim All Sectors 20% to 25%

F. Steingraber Manufacturing 25% to 30%

Service Sector 20% to 30%

A.V. Feigenbaum Manufacturing 15% to 40%

W.E. Deming Manufacturing 25% to 50%

Joseph M. Juran Manufacturing 20% to 40%

L.P. Carr Manufacturing 25% to 30%

Service Sector 30% to 50%

Average for All 22% to 35%

The Quality thought leaders have clearly identified the significant costs and losses in manufacturing and the service sectors when poor quality exists.

What is our experience of the Costs of Poor Quality Information? In the Costs of Poor Quality Information analyses that we have conducted, combined with the anecdotal evidence we have collected over the past twenty years, the evidence is clear. The Costs of Poor Quality Information as a percent of operating revenue or budget (for government and not-for-profit) is roughly equivalent to the costs of poor quality in the manufacturing and service sectors. In fact, manufacturing costs of poor quality often are caused by Information Quality issues such as errors in engineering and manufactur-ing specs or inaccurate inventory counts that cause materials shortages or materials and product overages.

In my first book Improving Data Warehouse and Business Information Quality, Wiley, (1999), I estimated the Costs of Poor Quality Information at from 10 to 25 percent of operating revenue. I was wrong!!! For the typical immature orga-nization, the costs are much higher.

34474c01.indd 21 9/1/09 8:39:48 AM


The costs of poor quality associated with the 122 anecdotes in Tables 1-1 and 1-2 totals US$1.2 Trillion, with several of these costs being annually recurring costs, see Figure 1-1.

The costs of Poor Quality Information — based on updated research and measure-ment — are from 20 to 35 percent of an organization’s operating revenue wasted in recovery from process failure and Information Scrap and Rework.

figure 1-1: The High Costs of Poor Quality Information

Information-intensive enterprises such as banks, insurance companies, gov-ernment agencies, and information brokers can have higher costs, reaching to 40% of operating revenue or budget or more.

Why Do We Have a Glut of Poor Quality Information and Information Processes?

There are many root causes, but a predominant cause is that business is man-aged out of the old, obsolete Industrial-Age principles of management by spe-cialization of labor:

Most organizations’ Leadership Teams have failed to heed Peter Drucker’s ■n

Information Age model of managing the Information-Based Organization19 as an orchestra ensemble (performing together as one entity) and W. Edwards Deming’s Pillar of the Enterprise as a (single) System.

Management creates performance measures and reward systems that ■n

compromise quality for speed and cost-reduction, failing to understand “Total Cost of Ownership.”

Western management tends not to go to ■n Gemba (the Japanese word for the Real Place where the Value Work takes place) to observe the information processes and find and fix the Root Causes of the broken processes. For more about Gemba, see the section “Solve Information Quality Problems at the Source — Not Downstream” in Chapter 2, “The ABCs of the TIQM Quality System for Total Information Quality Management.”

Management focuses on the needs of the investors over the needs of the ■n

end-Customers and Knowledge Workers who must delight the end-Customers.

Often, the truth of the matter is that Executive Management is out of the loop of feeling the pain of Poor Quality Information. Executives simply have to give the word to a staff member, “go get me this information,” and it is done. But the

34474c01.indd 22 9/1/09 8:39:48 AM


executive does not see how much hunting and chasing goes into finding and collating the information. On the other hand, Knowledge Workers who receive defective information are the ones who feel the pain. If management does not manage by walking around in Gemba (the “Real Place”) in which the value work goes on, they may never realize the enterprise-threatening costs they are incurring in process failure, Customer alienation, and Information Scrap and Rework to recover from the failure.

You cannot expect management to understand the real costs of poor qual-ity information until they go measure it in their own organization. Right now, leading-edge organizations that have matured their Information Quality com-petencies and have transformed their culture to a habit of continuous IQ process improvement are putting pressure on their competitors.

When management does measure the Costs of Poor Quality Information, they often find them so high as to dispute them as, “this cannot be true!!!” If they write these costs off, they put their organization in jeopardy of failure.

Organizations that do not understand the Taguchi Quality Loss Function (any variation from the nominal valve causes loss and waste) are in danger. This is true in manufacturing as well as Information Quality used to manage the business and manufacturing processes. The traditional measurement of costs of poor quality are warranty costs incurred by the manufacturer. Taguchi defines Quality Loss as “the loss imparted by the product to the society from the time the product is shipped.”20

How We Got into This Mess: The Root Causes of Business Failure in the Post-Industrial, Information Age

It should be well known that organizations cannot survive without having qual-ity information. But simply having processes that capture information does not guarantee that information captured, maintained, and delivered to Knowledge Workers has the quality to enable business effectiveness. In fact, to have consistent quality information requires an organization to apply the same kind of Quality Management Principles and processes to information as they have applied to manufacturing in order to produce quality products and services.

The irony of this is that Western business was precisely successful in applying the Industrial-Age principles to business. But when the paradigm shifted with the dawning of the Information Age, Business and Information Systems Management did not seem to recognize, understand, or embrace the paradigm shift. Business management failed to see that success in the emerging Information Age requires a fundamental rethinking of how it manages the enterprise. It was comfortable managing down the vertical functional organization lines, creating a spirit of competition rather than a spirit of cooperation, collaboration, and ensemble.

34474c01.indd 23 9/1/09 8:39:48 AM


Even when the U.S. re-discovered Dr. Deming in 1980 through an NBC docu-mentary film describing the transformation Japan made by applying sound principles of Quality Management, organizations jumped on the “TQM” band-wagon, without really understanding the fundamental principles and processes required. Many only paid lip service to Total Quality Management, without seeking to learn and understand. As a result the emphasis on TQM started falling away in most organizations, whose philosophy was quantity and speed, without understanding Costs of Ownership.

Where was Information Systems (IS) management? IS management seemed clueless as to the significance of database management technology and its uni-fying potential for managing Enterprise-Strength, Shared Information. Many IS leaders failed to see their work as a business function. This is clearly seen in the fact that many CIOs, by and large are NOT Chief Information Officers — but Chief Information Technology Officers. Falling into the techno-trap of believing their job was to put in place the information technology infrastructure, their job was then to build or acquire and deploy hardware, networks, and applications, period. Few CIOs saw and understood the Information-Age paradigm. Caught up in the ever-accelerating technology “innovation” spiral, they were not able to reason through what is the purpose — or need — for this or that technol-ogy. They could not see that their true purpose is to transform how work could be done in the realized Information Age, and that application “solutions” sic. “systems” is not their product — INFORMATION and empowered Knowledge Workers are.

The Information-Age purpose of the CIO has always been to deliver Quality, Just-In-Time Information to the Knowledge Workers, and transform how they perform work.

Avoiding the Crisis

You, the reader and your management, have three choices in addressing Information Quality:

1. You can maintain the status quo of your current rate of defective infor-mation, incurring the high costs of process failure and Information Scrap and Rework, possibly leading to risk of enterprise failure. You and your management cannot just write this off as, “but we’re different.” If you ignore the potential waste, it will only increase your losses.

2. You can embrace a conservative, reactive approach of “Playing at Quality” by implementing an “Inspect and Correct” model of data profiling and data “cleansing” [sic. “Information Scrap and Rework”]. This will also lead to competitive disadvantage if your competitors scoop your organization.

3. You can embrace the proactive model of Total Information Quality Management by designing quality and error-proofing into the processes.

34474c01.indd 24 9/1/09 8:39:48 AM


You can lead the transformation in your organization to the Stewardship that everyone in the organization is accountable for in their role and rela-tionship to Information and make IQ management a habit for everyone. In this model, your organization will create competitive advantage, and you will scoop your competitors.

I hope you embrace the third choice and help bring your organization into the realized Information Age.

Summary

The Costs of Poor Quality Information represent an incredible loss to society and all Stakeholders. The costs are not just economic, but they affect safety, well-being, and equitable treatment of all in the civilized societies around the world.

The total costs of poor quality Information and Information Systems in the 122 anecdotes represent nearly one and a quarter Trillion dollars of actual waste caused by poor quality Information and Information Systems, some of which are recurring annual costs.

The cost of a proactive IQ Management function is a fraction of the monies the IQ function will recover in Information Scrap and Rework as it creates the habit of Continuous Process Improvement.

As you read the following chapters and start applying the TIQM Quality System for Total Information Quality Management, I look forward to seeing you take your profits to the bank.

Remember that 122 major organizations have lost or cost others

$1,212,374,479,000 plus or minus a few $ Billion in Information Process failure.

While these costs are for mostly major organizations, no one is exempt from broken information processes. The more immature an organization is in applying Quality Management Principles, Processes, and Practices to its Information Processes, the higher the costs of recovery from process failure and Information Scrap and Rework it incurs.

Consider what your organization may be losing as a result of poor quality, Out-of-Control Information Processes. Then go find them, measure them, improve them and measure the return on your investment.

Then study Chapter 6, “The Step-by-Step Guide to Measuring the Costs and Risks of Poor Quality Information” and go measure these costs for the sake of your Enterprise and Enterprise Stakeholders. Remember, you have a stake in the survival of your Enterprise, as well.

34474c01.indd 25 9/1/09 8:39:48 AM

34474c01.indd 26 9/1/09 8:39:48 AM

from information crisis to business excellence through...

Documents