Computer Associates: Where Good Software Went to Die

How Charles Wang Built a $19 Billion Empire by Acquiring Companies, Firing Employees, and Milking Maintenance Revenue Until the End

There was a saying in the enterprise software industry during the 1990s and early 2000s: if Computer Associates acquired your company, update your resume. CA, founded in 1976 by Charles Wang and Russell Artzt, built itself into a mainframe software empire not through innovation but through an aggressive acquisition strategy that consumed over 200 companies across three decades.

The pattern was brutally consistent and widely known throughout the industry. CA would identify a successful software company with products that generated steady maintenance revenue, make an acquisition offer that couldn’t be refused, close the deal, and then immediately fire approximately 25 percent of the acquired company’s staff within days of Justice Department approval. The survivors would watch as CA put their beloved products into maintenance mode, extracting maximum cash flow while minimizing investment in development or customer support. As The Register memorably put it, CA acquired a reputation as the place decent software goes to die. This wasn’t an accident or an unfortunate side effect of growth strategy, it was the entire business model. CA became what Fortune described as the barnacle of corporate America: Once you had CA software onboard, it was so onerous and expensive to pull it out that few customers ever did. That led to a lot of steady cash flow and to arrogance on the part of CA’s management. The company’s customer base included nearly 100 percent of Fortune 1000 companies by the late 1990s, not because IT departments loved CA’s products but because replacing mission-critical mainframe software was prohibitively expensive and risky. You were trapped, and CA knew it, and the pricing reflected that captive audience dynamic in ways that made customers openly hostile to their vendor.

The acquisition binge started modestly in 1982 with Capex Corporation for $22 million, which was viewed as successful because it strengthened CA’s position in mainframe software for IBM’s OS operating system. But by 1987, CA was making moves that shocked the industry, acquiring rival Uccel for $830 million and temporarily becoming the largest independent software company ahead of Microsoft. Within five days of regulatory approval, CA dismissed 25 percent of Uccel’s 1,200 employees, establishing the cost-cutting playbook that would define every subsequent acquisition. The Uccel deal added 7,500 customers to CA’s base while eliminating a competitor, which was the dual purpose of nearly every CA acquisition: expand market share while reducing competition. By 1988, CA employed 4,500 people across 22 countries and reported sales of $842 million, servicing 30,000 customers including most Fortune 500 companies. But customer satisfaction ratings were abysmal because customers brought in by acquisitions complained of difficulty in finding the right person at CA to answer questions and of CA representatives who appeared to be more interested in sales than service.

The 1988 acquisition of Applied Data Research brought CA the widely-installed DATACOM/DB database management system, which became central to the Computing Architecture 1990s plan that Wang unveiled in April 1990 as an attempt to bring coherence to the chaotic collection of acquired products. The plan promised that all CA programs would work together and talk to each other across different hardware and operating systems, which sounded visionary but mostly served to justify the continuing acquisition spree.

The 1990s saw CA’s strategy reach its apex with mega-acquisitions that consolidated entire product categories under CA’s control. In 1995, Legent Corporation was acquired for $1.78 billion, the biggest-ever acquisition in the software industry at that time, and Cheyenne Software for $1.2 billion in 1996. Legent had itself been formed from a merger and brought robust enterprise systems management and client-server tools into CA’s portfolio, nearly doubling revenues in the following years but also requiring extensive product rationalization to eliminate redundancies. The 1999 purchase of Platinum Technology for $3.5 billion marked a pivotal expansion into Windows-based management and database tools, nearly doubling CA’s footprint in client-server software, though integration challenges proved substantial. Platinum brought with it products like ERwin data modeling technology (later sold off in 2016) and expanded CA’s reach beyond its traditional mainframe stronghold. The Sterling Software acquisition followed in 2000 for another multi-billion dollar deal, adding even more products to CA’s sprawling portfolio. But during the 1990s, Computer Associates cultivated a highly sales-oriented culture characterized by aggressive revenue targets and intense pressure on employees to meet quotas. This environment fostered a culture of fear, where arbitrary firings suppressed dissent, and decision-making was centralized among a small group of executives loyal to Wang, often prioritizing rapid growth through acquisitions and cost-cutting over formal processes. The late-quarter hockey stick deal-making became normalized, where sales teams would push to close massive contracts in the final days of quarters to hit Wall Street expectations, embedding aggressive tactics that would eventually lead to the company’s downfall.

The accounting scandal that erupted in the early 2000s exposed what many had suspected for years: CA’s revenue recognition practices were fraudulent. The company practiced what became known as the 35-day month, backdating contracts signed in the early days of a new quarter to count them as revenue in the previous quarter, allowing CA to meet Wall Street earnings expectations consistently through creative accounting rather than actual performance. Sales teams operated under a high-stakes model, with late-quarter hockey stick deal-making becoming normalized to hit Wall Street expectations, embedding aggressive tactics into daily operations. The accounting scandal of the early 2000s, involving fraudulent revenue recognition practices like the 35-day month, exposed deep cultural flaws and prompted significant reforms emphasizing ethics and accountability. CEO Sanjay Kumar, who had succeeded Charles Wang in 2000, resigned in April 2004 amid federal investigations and eventually served prison time for securities fraud and obstruction of justice. The scandal required CA to restate earnings going back to 1999, revealing that approximately $2.2 billion in revenue for fiscal years 2000 and 2001 had been prematurely recognized, with another $1.1 billion in earlier quarters improperly accounted for. Multiple executives went to prison. The company’s reputation, already poor due to its acquisition practices and customer service issues, was thoroughly destroyed. Yet remarkably, customers stayed because the products themselves were expensive and central to what corporate IT departments were doing, and so customers found it difficult to move away from CA. The switching costs for mainframe software were so high that even a company led by executives heading to federal prison maintained its customer base.

CA attempted rehabilitation under new leadership, bringing in IBM executive John Swainson as CEO in 2005 to restore stability and ethical practices. The company rebranded as CA Technologies, implemented corporate responsibility programs, and by the 2010s was being recognized by industry organizations for workplace quality and sustainability initiatives, suggesting genuine cultural reform. But the fundamental business model never changed: CA remained dependent on maintenance revenue from acquired products rather than organic innovation. The legacy database products acquired decades earlier, IDMS from Cullinet, DATACOM/DB from Applied Data Research, Ingres from ASK Group, continued generating steady revenue from organizations that couldn’t migrate away. CA Technologies had a reputation as the place where software companies with financial difficulties went to be bought out lest their products disappear completely. CA’s tactics, especially the deep downsizing of an acquired company’s staff, were often considered harsh. However, by continuing to support these products, CA provided a safety net for IT organizations and allowed them time to possibly migrate to alternative technologies.

This charitable interpretation suggests CA performed a valuable service by preventing software products from simply vanishing, though critics would argue that minimal support and declining investment hardly qualified as keeping products alive. In 2018, the story came full circle when CA Technologies itself was acquired by Broadcom for nearly $19 billion in cash, with the acquirer immediately implementing the same aggressive cost-cutting tactics that CA had pioneered. Broadcom laid off approximately 40 percent of CA employees in the United States, about 2,000 people, closed the Long Island headquarters that had been CA’s home for decades, and in January 2024, imploded the abandoned Islandia campus in a controlled demolition. The barnacle of corporate America was scraped off, its products absorbed into Broadcom’s portfolio where they continue generating maintenance revenue for legacy mainframe installations that will presumably run until the heat death of the universe. Charles Wang, who stepped down as CEO in 2000 and watched his successor go to prison while the company’s reputation cratered, had built something that lasted far longer than it probably deserved to. CA’s legacy isn’t the products it created (it created almost none) or the innovation it drove (negligible), it’s the business model it perfected: acquire established software with captive customers, minimize investment, maximize cash extraction, and repeat until either the market shifts or the accounting fraud gets discovered.

Every subsequent software roll-up strategy, every private equity firm buying enterprise software companies to milk maintenance revenue, every acquirer promising continued support while quietly planning layoffs, they’re all following the playbook CA wrote. The place where good software went to die taught an entire industry how profitable death could be.

Be Seeing You!

Kyle