presented its plans to bring
into its fold last week, the software company, not surprisingly, said it will be smooth sailing ahead.
But the plan has drawn its share of skeptics, who suggest choppy waters are more likely in the combined company's future. For one thing, PeopleSoft has a mixed track record in the M&A department. And in this case in particular, some analysts say, PeopleSoft's bold estimates for cross-selling products and cutting costs while maintaining three product lines add up to ambitious guidance that could prove disappointing next year. This, of course, assumes that
hostile bid for PeopleSoft flops -- the desire for which may have driven PeopleSoft's
Although some analysts ratcheted up their estimates to meet PeopleSoft's higher guidance for 2004, the consensus still falls short of the company's targets. Analysts polled by Thomson First Call are now forecasting the combined company will earn 85 cents a share in 2004 -- a nickel lower than the floor of the company's target range of 90 cents to 95 cents. The consensus estimate for 2004 revenue is $2.7 billion, also below the company's range of $2.8 billion to $2.9 billion.
"Over any medium or longer holding period, we believe that PSFT's aggressive guidance is likely to set the company up to disappoint, leading over time to negative revisions which should drive down the stock price," Sanford C. Bernstein analyst Charlie Di Bona said in a research note last week. Di Bona's earnings estimate of 77 cents a share is among the lowest on the Street. He has an underperform rating on PeopleSoft and his firm doesn't do investment banking business.
PeopleSoft has never embarked on an acquisition of the scale of the $1.9 billion J.D. Edwards deal, but its record with smaller acquisitions is mixed. The company has fared particularly badly in the supply-chain management sector, where it has made at least two acquisitions.
The first, in October 1996, was
Red Pepper Software
, at the time a hot supply-chain software maker on par with then privately held
. After the $225 million Red Pepper acquisition, PeopleSoft acquired
in September 1999 for $15.2 million.
Despite both acquisitions, PeopleSoft's supply-chain business has remained lackluster and rumors have swirled of the company making a splashier acquisition of
. PeopleSoft failed to make Gartner Dataquest's list of the top five vendors by supply-chain management license revenue in 2002. (
is at the top, followed by, i2, Oracle,
PeopleSoft spokesman Steve Swasey acknowledged the Red Pepper deal "was not as seamless an acquisition as it could have been." "It was not integrated as well as it could have been," he said. "They didn't integrate management teams. We didn't get a foothold as strongly in supply chain." But, he pointed out those acquisitions were done before CEO Craig Conway arrived.
PeopleSoft has stumbled in at least two other acquisitions, both orchestrated before Conway's arrival. PeopleSoft took exit charges totaling $34.6 million after failing to deliver products from its acquisitions of
, according to filings with the
Securities and Exchange Commission
. PeopleSoft acquired TriMark, which made insurance software, for $29.9 million and IPO-bound Intrepid, which made software for retailers, for $51.5 million.
When Conway came on board, one of his first moves was a major acquisition, of customer-relationship software maker
for $433 million. Since then, Conway's strategy has been to acquire smaller companies to tap their technologies to fill gaps in PeopleSoft products.
How well PeopleSoft integrated Vantive is debatable. In the company's analyst day presentation, Ram Gupta, executive vice president, boasted that PeopleSoft integrated Vantive into PeopleSoft's product line in six months.
While PeopleSoft was selling a CRM product based on Vantive's software within six months, the company didn't introduce a CRM product that was fully integrated with the rest of its suite for about 18 months. As PeopleSoft's Swasey explained, the company introduced its Internet-based PeopleSoft 8 human resources and financials software in July 2000, and then launched a new Internet-based PeopleSoft 8 customer-relationship management product in spring 2001, roughly 18 months after the Vantive acquisition closed at the end of 1999.
However, thanks in part to the Vantive acquisition, PeopleSoft has gained market share in the CRM market. The company has ranked No. 3 in the market for the past two years, but its market share has climbed to 4.3% in 2002 from 3.9% in 2001, according to Gartner Dataquest.
"The Vantive acquisition really was a very good example of how well PeopleSoft has done historically in terms of acquiring a significant customer base and maintaining them," said Josh Greenbaum, a principal with Enterprise Applications Consulting.
PeopleSoft is taking a different approach, however, with the J.D. Edwards deal. PeopleSoft is planning to sell three product lines: PeopleSoft Enterprise for large companies based on PeopleSoft products; PeopleSoft Enterprise One for the middle market, based on J.D. Edwards products; and PeopleSoft World, based on J.D. Edwards products for mainframe platforms.
Some analysts have said the company will have a difficult time maintaining three separate sales teams and three R&D teams. "Support for three separate code bases while building integration between PeopleSoft and J.D. Edwards product groups will require significant bandwidth and flawless execution," wrote Wachovia Securities analyst Kash Rangan in a research note. "In some cases, the integration effort will probably equate to building a new product suite." Rangan has a market perform rating on PeopleSoft and his firm hasn't done banking with PeopleSoft.
But other analysts say selling products based on three codes is no big deal. Jim Shepherd, a senior vice president at AMR Research, called it a "silly idea that software companies only can only have one product line and that a multibillion-dollar organization can't possibly manage more than one product line."
"I think this is one of those situations where despite the common wisdom that acquisitions are incredibly risky and rarely successful and fraught with problems, this is not one where I expect there to be big problems," Shepherd added.
However, the different code bases could limit the ability of sales teams to sell PeopleSoft products to J.D. Edwards customers. PeopleSoft has said it has identified $30 million to $40 million in revenue generated from cross-selling products to each other's customers. That may be hard to achieve because upwards of 70% of J.D. Edwards' installed base of customers are still using mainframes and therefore are less likely to buy PeopleSoft products, which operate on a different platform, Bernstein's Di Bona said.
In addition to forecasting revenue increases based on cross-selling products, PeopleSoft said cost cutting will help boost results in 2004. PeopleSoft forecast it will cut $167 million to $207 million in costs, reducing headcount by 7%, or 750 to 1,000 employees, to post an operating margin of at least 17% in 2004.
Di Bona also expressed doubts about those numbers, wondering if the company can meet its revenue goals if it cuts too much. He noted PeopleSoft plans to reduce R&D to 16% of sales vs. a current trailing 12-month R&D expenditure at 17.5% of sales.
Pacific Growth Equities analyst Patrick Mason raised concerns about the company meeting both revenue and cost-cutting goals in order to achieve a 17% operating margin. He noted that PeopleSoft has never achieved an operating margin above 13.5% for a full year, and its operating margin has only reached as high as 15.2% in any given quarter. Mason's model assumes a 15.5% operating margin, with revenue at $2.8 billion and earnings at 80 cents a share, at least a dime short of the company's guidance. Mason has an underweight rating on PeopleSoft and his firm hasn't done any business with the software firm.
"The company cited its experience in managing acquisitions," Mason wrote in a research note. "But when was the last time it acquired a 5,000-person organization that is two-thirds its size with a multitude of products to integrate across multiple verticals?"
The answer, Mason wrote: Never.