Edgar Filing: COMPUTER ASSOCIATES INTERNATIONAL INC

COMPUTER ASSOCIATES INT'L FORM DEFA14A

UNITED STATES
SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549

SCHEDULE 14A

Proxy Statement Pursuant to Section 14(a) of the Securities
Exchange Act of 1934 (Amendment No. )

Filed by the Registrant [X]
Filed by a Party other than the Registrant [ ]

Check the appropriate box:

[ ] Preliminary Proxy Statement
[ ] Confidential, for Use of the Commission Only (as permitted by Rule 14a-6(e)(2))
[ ] Definitive Proxy Statement
[ ] Definitive Additional Materials
[X] Soliciting Material Pursuant to § 240.14a-11(c) or § 240.14a-12

Computer Associates International, Inc.
(Name of Registrant as Specified In Its Charter)

(Name of Person(s) Filing Proxy Statement, if other than the Registrant)

Payment of Filing Fee (Check the appropriate box):
[X] No fee required.
[ ] Fee computed on table below per Exchange Act Rules 14a-6(i)(4) and 0-11.

1) Title of each class of securities to which transaction applies:
Not Applicable

2) Aggregate number of securities to which transaction applies:
Not Applicable

           3) Per unit price or other underlying value of transaction computed pursuant to Exchange
               Act Rule 0-11 (set forth the amount on which the filing fee is calculated and state how
               it was determined):
Not Applicable

4) Proposed maximum aggregate value of transaction:
Not Applicable

5) Total fee paid:
Not Applicable

[ ] Fee paid previously with preliminary materials.

[ ] Check box if any part of the fee is offset as provided by Exchange Act Rule 0-11(a)(2) and identify the filing for which the offsetting fee was paid previously. Identify the previous filing by registration statement number, or the Form or Schedule and the date of its filing.

1) Amount Previously Paid:
Not Applicable

2) Form, Schedule or Registration Statement No.:
Not Applicable

3) Filing Party:
Not Applicable

4) Date Filed:
Not Applicable

EXPLANATORY NOTE

Computer Associates International, Inc., a Delaware corporation ("Computer Associates"), is filing the materials contained in this Schedule 14A with the Securities and Exchange Commission on June 26, 2001 in connection with the solicitation of proxies for electing the board of directors of Computer Associates at the 2001 annual meeting of Computer Associates' stockholders.

[The Following Press Release was issued 06/26/01]

Contacts:	Computer Associates	Citigate Sard Verbinnen
	Bob Gordon	Owen Blicksilver/Jamie Tully,
	Public Relations	212-687-8080
	631-342-2391	OBlicksilver@sardverb.com /
	bobg@ca.com	JTully@sardverb.com

	Lisa Savino
	Investor Relations
	631-342-2788
	lisa.savino@ca.com

ComPuter associates challenges FINDINGS OF SURVEY commissioned BY Ranger Governance

Significant Flaws In Methodology and Interpretive Biases Uncovered;
CA Customer Satisfaction Levels Much Higher Than Wyly Study Purports To Show

CA Says Misleading, Self-Serving, Inflammatory Study
Only Hurts Employees and Shareholders

ISLANDIA, N.Y., June 26, 2001 - Computer Associates International, Inc. (NYSE: CA) today challenged the results of the "customer" survey commissioned by dissident Sam Wyly and used as a cornerstone in Wyly's campaign to elect a competing slate of directors to CA's board.

Said Sanjay Kumar, President and Chief Executive Officer of Computer Associates, "Wyly's assertions are based on fundamentally flawed and self-serving research that encompassed 76 interviews - 16 of which were with individuals who have never been customers of ours - and we flatly dispute the survey's conclusions. We grew to where we are today - serving tens of thousands of customers - by focusing intently on the needs of our customers. While good enough is never good enough, customer satisfaction is paramount to our success and we believe we have made significant strides in recent years, most recently with changes in our business model that respond directly to customer needs for increased choice and flexibility."

The Wyly study, which purports to show dissatisfaction among CA's customers, was reviewed by GuideStar Communications, Inc., a leading customer and employee satisfaction research firm that has been monitoring customer satisfaction for CA for the past two years.

In its report on the study, GuideStar said: "The methods of the poll as reported on the World Wide Web are not up to normally accepted standards for professional research, and the reporting of the data on the World Wide Web appears to be a thinly veiled work of propaganda, aimed at the unsuspecting lay reader, presented under the guise of a professional research study." Adding, "The margin of error for the Penn et al. study is, unfortunately, astronomical..."

Several critical methodology flaws and interpretive biases in the study were uncovered by GuideStar, including:

-	The fact only 76 individuals, out of a pool of tens of thousands of CA clients in North America, were interviewed, and 21% of the total sample (16 participants) explicitly stated that they had never used CA products. Moreover, there was no information on how the sample was chosen.

-	That questions were formulated in a manner to provoke negative responses, and

-	The high probability that negative comments on CA were over-weighted in the presentation of the study's findings on the Internet, while positive comments on CA were entirely omitted from the findings.

In addition, the study's conclusions are in dramatic contrast to the findings of GuideStar, which conducts ongoing in-depth, customer research for CA. In fact, a major CA customer survey conducted by GuideStar in early 2001 that included responses from a random sample of 2,204 respondents found that:

-	93.4% of respondents said that CA software was "important" to "mission-critical" to the successful operation of their business.

-	Fewer than 1 in 10 clients rated CA's service as poor.

-	Approximately 75% of CA's clients say they have a positive working relationship with CA.

Mr. Kumar continued: "CA understands the importance of frank and unbiased feedback from our customers, which is why CA has performed scientific customer surveys for many years. But Mr. Wyly's flawed, misleading and inflammatory study only hurts those who are truly committed to building on CA's success in the market - our employees and shareholders."

The GuideStar report, entitled Comparison Of CA 2001 North American Client Satisfaction Survey Findings To The Poll Findings Of Penn, Shoen, And Berland Associates, is available at: http://ca.com/media/guidestar_survey.pdf

IMPORTANT INFORMATION

Computer Associates plans to file a proxy statement with the Securities and Exchange Commission relating to Computer Associates' solicitation of proxies from the stockholders of Computer Associates with respect to the Computer Associates 2001 annual meeting of stockholders. COMPUTER ASSOCIATES ADVISES SECURITY HOLDERS TO READ ITS PROXY STATEMENT WHEN IT BECOMES AVAILABLE, BECAUSE IT WILL CONTAIN IMPORTANT INFORMATION. Computer Associates' proxy statement and other relevant documents will be available for free at www.sec.gov. You may also obtain a free copy of Computer Associates' proxy statement, when it becomes available, by writing to Computer Associates at One Computer Associates Plaza, Islandia, New York 11749, or at www.ca.com. Detailed information regarding the names, affiliation and interests of individuals who may be deemed participants in the solicitation of proxies of Computer Associates' shareholders is available in the soliciting materials on Schedule 14A filed by Computer Associates with the SEC.

About GuideStar Communications

GuideStar, formed in 1987, is a full-service, customer and employee satisfaction research firm with an integrated e-Survey Service Bureau providing quick, low-cost e-Surveys achieving robust response rates averaging 30% - 85%. GuideStar's mission is to find out the truth of things quickly and provide findings and counsel that assist our clients to measurably improve the productivity of their primary stakeholder relationships. GuideStar believes the best organizational survey research enhances people's relationships, connects with people's feelings, serves people's real needs, interests and concerns and plays a key role in moving organizations ahead. For additional information, please visit http://www.guidestarco.com.

About Computer Associates

Computer Associates International, Inc. (NYSE: CA) delivers The Software That Manages eBusiness. CA's world-class solutions address all aspects of eBusiness process management, information management, and infrastructure management in six focus areas: enterprise management, security, storage, eBusiness transformation and integration, portal and knowledge management, and predictive analysis and visualization. Founded in 1976, CA serves organizations in more than 100 countries, including 99 percent of the Fortune 500 companies. For more information, visit http://ca.com.

# # #

© 2001 Computer Associates International, Inc. One Computer Associates Plaza, Islandia, N.Y. 11749. All trademarks, trade names, service marks, and logos referenced herein belong to their respective companies.

[The Following Press Release was issued 06/26/01]

Contacts:	Computer Associates	Citigate Sard Verbinnen
	Bob Gordon	Owen Blicksilver/ Kimberly Polan
	Public Relations	212-687-8080
	631-342-2391	OBlicksilver@sardverb.com /
	bobg@ca.com	KPolan@sardverb.com

	Lisa Savino
	Investor Relations
	631-342-2788
	lisa.savino@ca.com

Computer Associates Names Two New Directors

Lewis Ranieri and Linus Cheung Bring Added Depth and Breadth To Board

ISLANDIA, N.Y., June 26, 2001 - Computer Associates International, Inc. (CA) today announced the appointment of two new members to its Board of Directors: Lewis S. Ranieri, Founder of Hyperion Partners L.P. and Chairman and CEO of Ranieri & Co. Inc., both private investment firms; and Linus W. L. Cheung, Deputy Chairman of Pacific Century CyberWorks and former CEO of Hong Kong Telecom.

"We are extremely pleased to welcome Lewis Ranieri and Linus Cheung to CA's Board of Directors," said Charles B. Wang, Chairman of Computer Associates. "Lew is one of the country's most accomplished financial executives and Linus brings extensive technology expertise and international businesses experience. These two highly successful men will add further breadth and depth to our Board of Directors. Their choice to join Computer Associates' team is a strong vote of confidence in our new business model and our growth prospects going forward."

"With its innovative technology, competitive new business model and commitment to first-class customer service, Computer Associates continues to demonstrate its industry leadership," said Lewis Ranieri. "I am very pleased to be joining the company at this dynamic time in its history and look forward to contributing to its continuing growth and success."

"I have always been impressed by Computer Associates' cutting-edge products and services, so I was particularly intrigued when Computer Associates approached me about joining the company's Board several months ago," said Linus Cheung. " I look forward to working closely with Computer Associates' exceptional management team and the rest of the Board of Directors as we seek to solidify the company's position as a leading international software provider."

Prior to forming Hyperion Partners, Mr. Ranieri, 54, was Vice Chairman of Salomon Brothers, Inc. Generally considered to be the "father" of the secondary mortgage market, Mr. Ranieri helped develop the capital markets as a source of funds for housing and commercial real estate. At Salomon Brothers, Mr. Ranieri had responsibility for all of the firm's activities in the mortgage, real estate and government-guaranteed areas. Mr. Ranieri has served on the National Association of Home Builders Mortgage Roundtable since 1986.

Before becoming Chief Executive Officer of Hong Kong Telecom, Mr. Cheung, 52, was Deputy Managing Director of Cathay Pacific Airways, where he worked for 23 years. Mr. Cheung was Executive Director of Cable & Wireless plc and served in the Central Policy Unit of the Hong Kong government during Governor Wilson's administration. Today, Mr. Cheung is a council member of the University of Hong Kong and the Hong Kong University of Science & Technology. He chairs the Management Board of the School of Business of the University of Hong Kong and sits on several government advisory committees.

IMPORTANT INFORMATION

About Computer Associates

# # #

[The Following was posted on CA's website on 06/26/01]

Comparison of CA 2001 North American Client
Satisfaction Survey Findings to the Poll Findings of
Penn, Shoen, and Berland Associates

Prepared by:

GuideStar Communications, Inc.

June 25, 2001

Purpose of the Comparison

A recent report posted on the Worldwide Web and discussed in the New York Times cited a survey of CA client satisfaction conducted by Penn, Shoen, and Berland Associates. The purpose of this brief report is to compare the methodologies and findings of the Penn et al. "opinion poll" to those of the most recent annual survey of client satisfaction conducted by GuideStar Communications in January to February of 2001.

GuideStar Communications has conducted the two most recent annual surveys of CA client satisfaction in North America. These included an initial, baseline study in early 2000 that used a random stratified sample of 1,638 CA product users and a follow-up, tracking study in early 2001 with a random stratified sample of 2,204 respondents. These two surveys form only one part of the ongoing, worldwide assessments of CA client satisfaction that we have been conducting over the past two years.

Our most recent North American Client Satisfaction Survey comprised more than 70, carefully developed items (questions) and had three alternate forms customized for different categories of users, including Chief Information Officers, IT Managers and End-users whose organizations have used CA products for an average of more than 10 years. We drew our sample from an exhaustive list of more than 11,000 CA clients representing more than1400 different companies in North America. We also have similar data, with similar findings, from Latin America, Europe, Australia, and New Zealand. Thus, we have a high degree of confidence in both our sampling methods and in the accuracy of our findings.

Statistically, the margin of error for our North American findings, overall, is less than 2%. Our respondents, all of whom are experienced CA clients, included 374 CIOs, 802 Managers and 1,028 End-users of CA products.

This report focuses on the data we obtained from the 2001 CA North American Client Satisfaction Survey. Additional data collected from CA client satisfaction surveys conducted internationally by GuideStar with CA clients in Europe, Latin America, Australia, and New Zealand, or from the 2000 Client Satisfaction Survey, are not included in this report.

Overall, we have three serious concerns, detailed below, with the findings presented on the Worldwide Web and attributed to Penn et al. These are:

	The clear inadequacy of the sample on which these conclusions were based.
-	The apparent biases with which their data have been interpreted.
-	The stark differences between their findings and ours.

[Chart setting forth flaws in the Penn et. al. "Poll"]

Samples

The Penn et al. poll was conducted with a sample of 76 individuals that included 52 interviews with current CA clients and 8 interviews with former CA clients. As can be seen on page 3 of their Web presentation (see www.rangergov.com), 45% of these interviewees were Chief Information or Technical Officers; the remainder were corporate executives outside of the IT department. Based on our extensive experience, these individuals are unlikely to have any hands-on experience with CA software products. There were no IT Managers or End-Users. These are the people within CA client IT organizations that do make extensive use of CA products. Thus, it appears that few clients with actual current hands-on experience as users of CA products were interviewed.

Sixteen interviews (21% of the total sample) were conducted with individuals who explicitly state that they have never used CA products. It is unclear what relevance these individuals have to client satisfaction.

It is unclear whether most, or how many, of the survey respondents have any personal experience with the use of CA products. The relevance of the Penn et al poll sample to client satisfaction is questionable, at best.

[Chart setting forth comparison of numbers of CA product users interviewed in the Penn and GuideStar studies]

Perhaps the single most important factor in determining the validity and utility of survey data is the degree to which the sample obtained is representative of the population to which one wishes to generalize the findings. That is, can we use the opinions that we obtain with our poll or survey to make general statements? At least two factors contribute to our confidence in doing so. The first is that the sample must be randomly obtained from the parent population. That is, we need a random sample of CA clients. An obvious preference would be to limit our sample to individuals who are directly involved with CA product use.

Second, we need to know that the number of people we poll or survey is sufficient to give us a fairly accurate estimation of the percentage of respondents who give one opinion or another (or one average score or another). For example, if 50% of respondents say that they like a particular product, we must ask how accurate that estimate is likely to be when projected onto the entire population represented by the sample in the survey.

Here we can use statistical procedures that, given the size of the obtained sample, tell us about the confidence that we can have in our estimates when projecting the findings of the survey sample onto the larger population as a whole. In political polls, therefore, you will often see such results presented, for example, as being 50% ± 4 percentage points. That is a way of saying that the true percentage is likely to be somewhere between 46% and 54%.

The larger your sample size in relation to the size of the population you draw it from, the more accurate your estimate is likely to be. Conversely, the smaller your sample size is compared with the size of the population you draw it from, the more inaccurate your estimate is likely to be.

[Chart setting forth a comparison of the quality of research in the Penn and GuideStar studies]

At best, the Penn et al. poll provides information on 76 individuals associated with no more, and perhaps less, than 76 companies. CA, however, is among the top four software companies in the world with approximately tens of thousands clients in North America. A sample that could possibly represent, at most, 60 companies that use or have used CA software is inadequate to provide accurate estimates of the percentage of clients who like, dislike, or have any other opinion of CA software and products. In essence, the sample size is so small that the data are anecdotal at best. That is not to say that all opinions are not meaningful, only that the Penn et al. sample is too small to give us accurate information on CA product users in general.

Unfortunately, Penn et al. provide no information on how their sample was selected. We can assume with some certainty, however, that it does not represent a random sample of CA product users. We assume this, quite frankly, because Penn et al. do not have access to a list of CA users. Even if they had such access, however, the sample size would still be inadequate to draw any meaningful conclusions.

The CA 2001 Client Satisfaction Survey sample conducted by GuideStar Communications was a random sample of CA clients that included Chief Information and Technical Officers, relevant Managers, and End-Users selected from a current listing of over 11,000 current CA clients. As stated earlier, a total of 2,204 respondents from 1,400 different organizations (corporations plus federal, state and local governments, educational and non-profit institutions) provided us with feedback on CA products and services. Using a 95% confidence interval, our estimates were computed to be accurate to within 1.88%.

The GuideStar survey sample is of sufficient size to be able to state, mathematically (95% confidence interval), the range in which the true values for each of our estimates are likely to be. Thus, if we say that "50% of respondents" feel one way or another, then we can estimate that the true value is likely to be within the range of 48-52%. The margin of error for the Penn et al. study is, unfortunately, astronomical in comparison. The survey sample that they obtained is simply too small for their intended, or hoped for, purpose in conducting the study.

Their study as presented suffers from a number of additional disadvantages in comparison to ours. For example, we have findings from two consecutive years, from many countries, across multiple continents. And, we have access to the CA client database. We can and did, therefore, invite truly random samples to participate in each of the surveys that we conducted for CA. And, we can and did compare the data we obtained with these different samples and at two different times.

In essence, our findings and recommendations to CA were replicated (found and found again) on multiple occasions, with many users, across regions, cultures, languages, and products. Despite the limitations of survey methods in general, we have confidence in the quantity and the quality of the data we obtained.

Comparison of Findings

In all fairness, we do not know what the stated mission or actual interpretations of Penn et al. were. We are only aware of how the press has interpreted these data and how they were summarized on the Worldwide Web. Nevertheless, we feel compelled to point out that these interpretations stand in stark contrast to our findings. Here are just a few important points to note:

-	The sample and sampling method of the Penn et al. "poll" do not reflect the rigor, size, or quality of the data that GuideStar has, through the resources and ongoing commitments of CA, been able to apply to our findings.
-	We do not believe, from a methodological viewpoint, that the data obtained by Penn et al. can be interpreted with confidence.
-	The methods of the poll as reported on the Worldwide Web are not up to normally accepted standards for professional survey research, and the reporting of the data on the Worldwide Web appears to be a thinly veiled work of propaganda aimed at the unsuspecting lay reader presented under the guise of a professional research study.
-	We find on examination that many of the summary findings reported on the Worldwide Web and in the NY Times are not in agreement with the findings of GuideStar's more extensive, independent research on CA client satisfaction.

Without belaboring the points above, here are some key examples:

[Chart setting forth the overall views of CA clients reported in the GuideStar study]

Penn, Shoen, and Berland, suggested that more clients are dissatisfied than are satisfied with their relationship with CA. In a clear example cited in the NY Times, however, this is phrased as comparing the percentage that are "dissatisfied" (41%) to those who are "highly satisfied" (6%). This appears to be a rather "loaded" comparison in which clients who are merely "satisfied" were excluded but those who were either "dissatisfied" or "highly dissatisfied" were grouped into a single category. No matter how you combine categories, however, our findings are quite different. As shown in the graph above:

*	GuideStar found that approximately 75% of CA clients say they have a positive working relationship with CA.

*	93% of respondents said that CA software was "important" to "mission critical" to the successful operation of their business.

*	Fewer than 1 in 10 clients rated CA's service as poor. The majority (78%) rated it as "average" to "excellent."

*	Most (67%) clients in the 2001 CA Client Satisfaction Survey rated CA's overall service performance as equal to or better than that of their major competitors.

As to the CA products themselves:

-	Our findings show that 72% said they were "satisfied" to "very satisfied" with the quality of CA products.

Perhaps more disturbing is the consistent theme within the Web posting of focusing on negative numbers. For example, various figures of 12% to 33% of customers are reported as being "dissatisfied" with how CA meets their needs, with the sales process, with CA's responsiveness to requests, and with the value that they get from CA products and services. All negative figures. What of the 67% to 88% of respondents, the clear majority, whose opinions are not presented?

The authors repeat this pattern in reporting their findings on "customers" opinions of how senior management perform, citing figures of dissatisfaction or distrust of 21-33%. Again, the opinions of the majority of the sample are not reported. Should not a satisfaction study include a report of what all respondents, including satisfied customers, who appear to represent the majority even in the Penn et al. data, have to say? Certainly, more balance in the presentation is warranted.

Often it appears that the authors of this summary, whomever they may be, are putting a great deal of effort into trying to put negative spins on their findings. The "major findings" presented on pages 25-30 are informative examples. Those interviewed are asked if they have read any recent press reports about CA accounting practices, etc. "Few" have. In actuality, this few is 27 people. Of those, "about half" (13-14 people?) mention having read about "accounting and customer/employee coverage." Of those, 18% (2-3 people?) said it was either "negative" or they have "question marks about them." The next most frequent category (12%; 1-2 people?) recall something about a "change of image/logo." All of the rest of the categories represent about 6% of the "about half" of the people who were interviewed; this equals one or perhaps two people. Shall we generalize from him or her to tens of thousands of CA North American clients?

Having reported the above, they then go on to ask if whether what the interviewees recall having read would influence their impressions of CA. More than half (61%) said it would not make any difference and 17% said they would view CA more positively.

From there, it goes on. Those interviewed are then told about "allegations that CA are not responsive and treat their clients poorly." This is perhaps one of the boldest and clearest instances of attempting to bias a survey that one could ever obtain. Indeed, it is a textbook example. But, quite obviously, having now been told this by the interviewer, most clients are indeed "concerned" about such "allegations." But, even after then going on to tell those interviewed that "there was press coverage calling into question CA's accounting practices," 75% of respondents still said they would be "somewhat" to "very likely" do business with CA.

Are these mere examples of our abstracting out trivial sections of the Web report for condemnation? We think not. Rather, we believe these represent a recurrent theme of negative biasing. All of the comments made by those interviewed that are presented in the report are negative. Were there no positive comments? And why do those to whom the comments are attributed seem to appear over and over again?

A "CIO/Manufacturing" is mentioned 8 times in the report. Are these 8 comments by the same person or 8 different individuals? Our reading suggests that at least some of these comments are probably multiple comments from single individuals. Why, do we believe this? Because the comments are compiled from less than half the sample (41%) and the most common category of comments (price) included comments made by only 8 people (one in four of the 41% of the 76 individuals interviewed equals 8 people).

The vociferous criticisms mentioned in the summary posted on the Web, according to their categorical breakdown in the table on page 15, would have had to have come from 5 or fewer people. Thus, the 8 comments mentioned above are likely to have come from fewer than 8, and possibly only one person, who was interviewed for the poll.

We have thousands of open-ended comments in our database. We have coded, analyzed, and summarized all of them. Many offer very constructive suggestions for improving CA products and services. For balance, we have appended the first 100 positive, unedited comments obtained in response to one of our questions. At least now readers can get some idea of things that clients like about CA.

In total, our written summary report to CA on the North American Client Satisfaction survey was more than 50 single-space pages in length. The above represent merely a sampling of the more important discrepancies between the Penn et al. findings and ours.

Overall, we find a disturbing tendency of the report on the Worldwide Web to abstract out the most negative aspects of the findings and the most negative comments. Any attempts at a balanced presentation of the data appear to have been lost in their posting.

Here are a few more clear examples:

-	The report of Penn et al's findings notes that 1 in 5 of those surveyed who have never been CA clients have "an unfavorable opinion of the company." Would it be true to say, then, that 80% of non-clients think highly of CA? If so, this represents a tremendous market potential for their products! Unfortunately, the presentation on the Web is too sketchy to know.

-	The report also highlights the "troubling sign for CA" that 36% of their clients "would not recommend" CA to someone else. Conversely, their graph shows that a majority (61%) would. Why the constant focus on negative findings, regardless of their proportion in the sample? Are these data being interpreted solely to make CA look bad in the eyes of the readers?

The purpose of client satisfaction surveys is to identify the strengths and weaknesses of a company's relationships with their clients. All companies have weaknesses and areas in need of remedial attention. Ideally, the results of client surveys should serve to guide companies along the path towards excellence.

CA, like all companies, has both strengths and weaknesses in their client relationships. We have and continue to provide them with carefully obtained and interpreted data to assist them in their ongoing efforts to provide better and better service. Our data suggest that improvements are occurring. Between 2000 and 2001, for example, CA increased the size of its Client Relationship Organization. Client contacts with their Client Relationship Managers (CRM) are increasing as a result. Early findings show increased CRM contacts are associated with increased client satisfaction.

Indeed, we have obtained data showing increases of 10-15% in satisfaction across a number of areas over the past year. Some were even greater.

Thirty percent more clients reported improvements in client service, 15% more increased their satisfaction with CA's responsiveness, 14% more were satisfied with how their service requests were resolved, 13% more rated CA as a "good" to "excellent" strategic partner, 10% more rated CA as "good" to "excellent" in understanding their IT needs, 12% more rated them as "good" to "excellent" in understanding their business needs, etc.

The Web posting reported that only one-fifth of the people that Penn et al. interviewed were aware of the increased CRO staffing. We found that 47% had already felt its effects by the time of our latest survey 6 months ago and they indicated that it had positively impacted on their ease of doing business with CA. More importantly, among clients who had begun receiving more frequent CRM in-person contacts at that time, we observed a dramatic 42% increase in their stated likelihood of recommending CA to other organizations. In short, our findings in a much larger and more representative sample are quite different from those touted on the Worldwide Web.

But we are not a public relations company. We are an independent, corporate research consultant firm, and we feel compelled to issue our concerns about how data on a small, inadequate sample appear to have been used as a basis to attack a company for which we have obtained better, and contradictory, data on client satisfaction.

Ira Kerns
Principal Consultant

David E. Schotte, Ph.D.
Senior Research Analyst

GuideStar Communications, Inc.
www.guidestarco.com

Appendix

Positive Comments on CA - The First 100
(Abstracted, verbatim and in order of appearance)

	If you were going to recommend CA to another organization, which qualities or strengths of CA would influence you to recommend CA?
1.	Ability to understand our organization's needs
2.	Ability to present creative cost solutions
3.	The partnership and attention from the sales staff has been good, even though I have dealt with multiple sales reps.
4.	This depends entirely upon the product. I would very likely recommend certain CA products and not others.
5.	Major player in the industry.
6.	Solutions provider for enterprise.
7.	Scope of products and technical strengths.
8.	large variety of products
9.	RELIABLE PRODUCTS
10.	RELIABLE PRODUCTS
11.	We are predominately a mainframe shop and have bees successful with Datacon/db, I like theproduct very much and would recommend it. The one drawback is the continued increase in cost as our business continues to grow as does our computing environment. The people I deal with are great. The tech area provides excellent value. Your organization also seems very willing to learn our business as well.
12	Broad range of products Good people
13.	Products
14.	Customer Service
15.	Quality of the product
16.	Good quality products, good tech support.
17.	Product selection, the ability to assess a company's needs and gauge recommendations appropriately.
18.	Product range
19.	Comprehensive solutions cost effective partnering
20.	They have such a big presence, particularly in the mainframe world, that it is hard to ignore them.
21.	We are currently only maintaining legacy applications and have not stayed current with many of CA's offerings.
22.	Solid products
23.	Good product, at least platinum Fastunload product.
24.	Do not recommend based on the company...CA or others; would do a technical brief of use. Sales exec experience influences this to some extent also
25.	Database products are excellent for our environment
26.	Technical Support Customer Service
27.	Comprehensive product line
28	Software stability; you do get what the software literature says you'll get; almost every vendor is putting out software before it's really been tested and CA is a little better than others - especially IBM and Tivoli who apparently feels that the customers ought to test for them!
29.	Quality, service
30.	Tough negotiators, but fair High integrity Knowledgeable
31.	Size, breadth of products
32.	Experience with all areas of ESM
33.	A company who can enterprise IT solutions
34.	Overall strength in the marketplace for products to support large data centers with diverse operating systems
35.	High performance RDBMS
36.	Extremely professional, technical, and sound understanding of where technology is going today and in the future.
37.	AHD is an excellent product, Agent technology is good, Security is good as is RCO. Presales is a definite strength.
38.	Product quality and support
39.	Product quality
40.	Good product and support
41.	EDS and CA relationship
42.	Good tech support and quality of product is good
43.	Quality of CA's software. It is solid.
44.	Integrity, CA Ownership and responsibility for their products and their performance, reliability, Senior Executive involvement with customers.
45	Innovative toolset
46.	Product quality and service
47.	Breadth and depth of organization, global presence. It is a company where a good rep can get a lot done for a customer quickly.
48.	QUALITY OF PRODUCTS.
49.	Quality of the products and potential business process impact.
50.	Product scope and functionality Strong Top Level Management Appear to be moving toward more customer oriented/centered approach, which includes more the Burger King approach versus My way or the Highway Approach (as has been in the past).
51.	Ability of the tools to integrate reasonably well in a standard operating environment.
52.	I like the product but I need them to pull the installation off.
53.	Size of the organization coupled with strategic direction
54.	Products
55.	Strength is the size of the Company and their indicated interest in being more proactive.
56.	Software technical qualities. Good Pre-sales Technical support.
57.	Product and service strengths
58.	I HAVE BEEN IMPRESSED WITH THE CHANGE IN THE PROFESSIONALISM AND WORKING RELATIONSHIP WITH THE CA SALES REPRESENTATIVES AND THE INTRODUCTION OF THE CRM POSTION. THIS HAS MADE DEALING WITH CA FAR MORE PRODUCTIVE AND A MORE POSITIVE EXPERIENCE. THE OVERALL CONTINUITY OF THE ACCOUNT TEAM HAS PLAYED A BIG PERT IN THE RELATIONSHIP BETWEEN US.
59.	Existing Software
60.	Strong market share and stability of organization.
61.	The number one advantage of CA has been the products are useful to our organization and are fairly reliable.
62.	Commitment of local phx (az) acct team
63.	Individual product strengths
64.	Focus and technical support
65	The technical team and consultants I have worked with have been very knowledgeable and effective in resolving issues.
66.	Product categories and strength of products
67.	Depends on the products involved.
68.	Software quality
69.	Products work; mission critical products are good. We have a good relationship with sales team.
70.	CA has some very solid products depending on operational requirements. Strength of product would be my biggest sales point.
71.	Quality
72.	Breadth of products
73.	Quality, but costly products
74.	Has some good to excellent products
75	I do believe that CA has a good product. The functionality of the product is your biggest plus.
76.	Product selection
77.	Good products.
78.	Responsive to help any outstanding issues.
79.	Have quality existing products.
80.	CA is attempting to change its image, and trying to become more customer oriented. Also becoming more creative in pricing.
81.	Consistency of message. Excellent phone support
82.	Tech Support very good. Most CA products we use are legacy from other companies, i.e. Legent, Uccel
83.	For specific products and services that might be compatible to others. It will be based on experience with a product or service.
84.	Range of product and services options. Depth of resources.
85	Breadth of offerings
86.	Strong product line. Good, responsive people.
87.	CA has many excellent software products that perform well.
88.	Client Service
89.	Easy to do business with. Has my interests at heart and not just CA. Should be more of a win/win and the ability to lose with class.
90.	Leading edge of product development and cultural bias to listening to customer's ideas and needs.
91.	Functionality of product suite.
92.	Specific products that we have found benefit
93.	Product functionality, technical support.
94.	Relationship with Charles Wang and product that works to provide value to systems infrastructure
95.	It appears that when we get your product adjusted to our organization, it will be very beneficial.
96.	The new client relationship team is a positive step in building strong relationships between CA and its customers.
97.	Strong technical support. Innovative sales ideas, personal senior level involvement in your problems

IMPORTANT INFORMATION

INFORMATION REGARDING PARTICIPANTS

Computer Associates International, Inc. ("Computer Associates") and the directors of Computer Associates may be deemed to be participants in the solicitation of proxies (the "Solicitation") in respect of electing the board of directors of Computer Associates at the 2001 annual meeting of stockholders of Computer Associates. The current directors of Computer Associates (the "Directors") are: Russell M. Artzt, Alfonse M. D'Amato, Willem F.P. de Vogel, Richard A. Grasso, Shirley Strum Kenny, Sanjay Kumar, Roel Pieper and Charles B. Wang.

The Directors have interests in the Solicitation from their beneficial ownership of the common stock of Computer Associates. The Directors receive customary compensation from Computer Associates in exchange for their services as directors. Additional information with respect to the beneficial ownership of the Directors of Computer Associates common stock is set forth immediately below:

Name of Director	Computer Associates Common Shares
Russell M. Artzt	2,461,201
Linus Cheung	0
Alfonse M. D'Amato	13,500
Willem F.P. de Vogel	71,052
Richard A. Grasso	62,250
Shirley Strum Kenny	13,000
Sanjay Kumar	4,451,986
Roel Pieper	13,500
Lewis Ranieri	0
Charles B. Wang	29,732,991