On IBM buying Cognos

November 18, 2007


I.B.M. Acquires Cognos, Maker of Business Software, for $4.9 Billion

Cognos, the last major independent producer of software that companies use to analyze mountains of data, agreed on Monday to be acquired by I.B.M. for $4.9 billion in cash.

The takeover of Cognos, which is based in Ottawa, followed last month’s acquisition of the French company Business Objects, another maker of business intelligence software, by SAP of Germany for $6.8 billion. In March, Oracle bought Hyperion, another competitor of Cognos.

Software that sifts through data to tease out things like a customer’s buying habits or a corporation’s inefficiencies is hardly new. Cognos itself was founded in 1969. But several factors have prompted the recent interest in this market segment from larger software companies and led to Cognos’s relatively high price tag.

If the deal is approved by Cognos’s shareholders, I.B.M. will pay about 5 times the company’s annual revenue and 39 times its annual net earnings of about $126 million. Takeover speculation has driven the value of Cognos shares up in recent weeks. The I.B.M. bid of $58 a share represents a 9.5 percent premium to its Friday closing price.

But many analysts anticipate that the business intelligence software market is about to enter a period of significant demand from companies and government agencies.

“I.B.M. had nothing in this area,” said Colleen Graham, research director at Gartner, a technology consulting firm. “I.B.M. had to do something. But because it didn’t move fast enough, it paid a premium.”

The acquisition of Cognos will present I.B.M. with some unusual integration issues. Business intelligence software has traditionally been intended to be compatible with a variety of computer applications from a wide number of software companies.

Doing that requires cooperation from those other companies on technical issues. But Sanju K. Bansal, the chief operating officer of MicroStrategy, a relatively small maker of business intelligence software, anticipates Cognos may find those companies less helpful after it becomes part of I.B.M.

“Those vendors don’t want to cooperate with I.B.M. because they compete with I.B.M. every day,” he said.

There may also be conflicts with companies like Accenture that include Cognos software when they sell large data systems to corporations. Those firms are also direct competitors to I.B.M. in many areas.

But Bob Djurdjevic, president of Annex Research, a consulting firm, said Cognos would find it relatively easy to recover any lost business once it is part of I.B.M.

“Cognos will have a distribution system it could only have dreamed of,” Mr. Djurdjevic said.

In a conference call with reporters, Rob Ashe, the president and chief executive of Cognos, said he would continue to run the company from Ottawa. While I.B.M. intends to maintain the Cognos brand name, the company will be an I.B.M. unit rather than a stand-alone entity.

Like most business intelligence companies, Cognos software has a following among many retailers including Home Depot, Amazon.com, American Eagle Outfitters and 7-Eleven. Harrah’s, the casino operator, similarly uses loyalty programs and Cognos software to tailor incentives for gamblers.

Ms. Graham at Gartner said that the stricter financial reporting requirements imposed by the Sarbanes-Oxley Act were having the unintended side effect of increasing interest in business intelligence software.

Meeting the new regulatory requirements has forced many companies to clean up their computer systems and databases, a process that often involves purchasing large packages of programs from companies like SAP, Oracle and I.B.M.

“When they go through all that effort, many of them consider adding B.I. software to see if they can get some additional cost benefits from the process,” Ms. Graham said.

News from other sources:

IBM Takes Cognos for Next-Gen Business Intelligence

IBM said the acquisition of Cognos will help it deliver the “next generation” of BI (business intelligence) tools that customers will require as markets become increasingly global and competitive.

More specifically, IBM senior vice president Steve Mills said that Cognos technology is focused on processing information innear-real-time, rather than retrospectively.

Cognos BI applications are consistent with “a real-time, prospective approach to business analysis,” Mills said during a conference call with reporters to discuss the acquisition.

Cognos and IBM have partnered for more than 15 years, but Mills said that the acquisition will help IBM develop applications that customers will want in the future.

In an answer to a question from eWEEK, Mills noted that “the quantity of data [customers are] going to want to analyze is going up.” He asserted that independent companies are less able to produce “the next generation of business intelligence capabilities” that customers require.

He cited previous IBM acquisitions of Ascential and Rational as examples of where IBM acquired business partners to deliver greater value to the market.

Cognos is the 23rd IBM acquisition in support of its Information on Demand strategy, which is intended to help customers deliver more business insights to a broader set of people across an organization.

Other strategic acquisitions in support of IBM’s Information on Demand initiative include Princeton Softech (data archiving and compliance), FileNet (enterprise content management), Ascential Software (information integration), DataMirror (changed data capture), SRD (entity analytics), Trigo (product information management), DWL (customer information management) and Alphablox (analytics).

IBM has identified BI as an important growing sector of the enterprise software space. Using numbers from IDC, the company said BI Tools represent a $7.8 billion opportunity in 2008, growing at almost 12 percent per year. It added that the overall market opportunity for BI software and services is expected to be $30.6 billion for 2008.

IBM‘s $5 Billion Cognos Offer Signals Shift Beyond Middleware

Cognos offers business intelligence tools, like its Cognos 8 suite, that let users generate reports, charts, and other tools through which they can quickly get a handle on metrics such as sales, inventories, fulfillment rates, and the like. The Ottawa-based company’s customers include corporate giants like BMW, Bank of America and Dow Chemical.

For its part, IBM — in part through acquisitions — has been building a strong portfolio of middleware that can extract raw data from various departments within an enterprise and feed it to business intelligence tools like those offered by Cognos.

In recent months, IBM has bought out middleware vendors Ascential Software, FileNet, Trigo, and others.

By adding Cognos, IBM would be in a position to offer customers a complete information management platform that begins with its DB2 database line, runs through its middleware offerings, and extends out to Cognos-powered dashboards and reports tapped by line-of-business end users.

The battle for business intelligence

Business-intelligence software allows companies to mine and turn overwhelming amounts of data into coherent information that could be used for a wide range of business tasks, such as monitoring sales or keeping track of inventory.

The market’s grown in response to the explosion of information that’s ensued as businesses began producing massive amounts of data over the last decade.

“They are producing so much data and they are inundated with it,” said Gary Beach, publisher of CIO Magazine, which is geared toward chief information officers. “Business intelligence is a solution to massage that data for business purposes rather than have it sit in silos.”

“It’s like Google Maps for business data,” he added. “It allows you to monitor data to minutiae levels. If you work for Wal-Mart, you can drill down on a region, a store, an aisle.

Smaller, pure-play software companies, such as Cognos and Hyperion Solutions Corp., had been the key players in the business-intelligence market. But as the software became a more critical tool in the business world, the big guns have taken notice.

Indeed, the IBM deal is the third major acquisition of a business-intelligence company over the past eight months.

In March, Oracle Corp announced a plan to buy Hyperion for $3.3 billion, a move that analysts predicted could trigger a domino effect.

That prediction is proving accurate, as borne out by last month’s announcement by Oracle rival SAP for $6.8 billion — and now by the IBM deal.

The large vendors are looking to take over the business-intelligence territory and carve it between themselves,” said Colleen Graham of research firm Gartner.

“Tell me your BI plan”

Beach said the tech giants have to have a business-intelligence tool as part of their portfolio in order to be competitive.

“They need that arrow in their quiver to put in front of customers,” he said. “Customers are smart. They’ll say, ‘Tell me your BI plan.’ They need a direct answer to that.”

Graham cited the example of Oracle. Chief Executive Larry Ellison has been making good on his vow to lead the consolidation of the corporate software market.

“Oracle clearly wants to own the whole stack,” she said. But until it purchased Hyperion, she added, Oracle didn’t have much to show in terms of business-intelligence tools.

Similarly, Graham added, IBM’s decision to buy Canada’s Cognos is also a move to address a “competitive hole” in Big Blue’s corporate software portfolio.

Beach said the recent deals underscore the difficulty that smaller software companies are having competing in business intelligence.

“It’s getting increasingly difficult for best-of-breed vendors to get a seat at the table,” he said.

But Russ Cobb, a vice president at SAS, one of the remaining stand-alone players in the business intelligence market, said the recent acquisitions of its rivals have created opportunities for the Cary, N.C.-based company, which also offers data integration and other business-software products.

He said that the Hyperion and Business Objects acquisitions have led to periods of distraction for the acquired companies, and he expects the same thing to happen with the IBM deal. “We expect this one to be similar to the other two, giving us opportunities in the market,” he said.

IBM to buy Canada’s Cognos for $5 billion

“This is the end of business intelligence as we know it. The IBM move is evidence of the domino effect that’s affecting the sector,” he said.

IBM has long said it wanted to develop its business-intelligence products internally, but the latest deal is the 23rd acquisition over the past year and a half to support its so-called Information on Demand strategy. Analyst Boris Evelson of Forrester Research said the Cognos deal shows that “organic growth is not happening fast enough.”

The goal of the strategy, first unveiled in February 2006, is to provide a complete suite of software to help companies manage heir data and run their businesses more efficiently. Cognos helps fill a major gap in IBM’s portfolio.

The acquisition is expected to close in the first quarter of 2008 after a “plan of arrangement” is completed. Under that type of merger approval, a special meeting of shareholders is set up and two-thirds of the votes cast must favor the transaction.

Largest IBM Deal Ever: $5B for Cognos

IBM Corp. is making its largest acquisition ever, a deal announced Monday to buy Cognos Inc. for $5 billion in cash in hopes of keeping up with rivals in “business intelligence” software.

The acquisition would follow similar moves in the same market this year. SAP AG recently linked up with Business Objects SA for $7 billion and Oracle Corp. grabbed Hyperion Solutions Corp. for $3.3 billion. Cognos shares had soared recently on expectations that it, too, would be acquired. They leaped another 8 percent on Monday’s news.

Business-intelligence software helps big organizations gather data in “dashboards” that can be used to model such things as the financial impact of staffing changes or marketing moves.

Because that capability meshes with other things business software does, big vendors like SAP, Oracle and IBM have moved to include business-intelligence applications in their product packages.

But their challenge will be to technically integrate business-intelligence offerings in a way that is easier and less expensive for customers. David O’Connell, a senior analyst with Nucleus Research, said he was skeptical that will happen with IBM and Cognos.

“I’m not sure it will become more than a bolt-on,” he said.

Forrester Research’s Paul Hamerman agreed that it could take time for IBM to produce compelling technological linkages, but he added, “I don’t see any downside for customers.” IBM and Cognos already had a business partnership.

IBM has been on an acquisition tear in recent years to build out its software portfolio, because software generates much fatter profit margins than IBM’s sprawling technology-services business. IBM had been primarily a provider of “middleware,” which connects various kinds of software applications, but its recent push has led the company to expand more into the business of selling applications as well.

Cognos will be added to IBM’s information management software division. Cognos CEO Rob Ashe is expected to remain and report to the group’s head, Ambuj Goyal, after the deal is completed in the first quarter of 2008.


Cognos Buys More Pipe for IBM

Guy Creese, an analyst with Burton Group, said IBM’s acquisition of Cognos fits in with IBM’s information management initiative and “they get a company with $1 billion in revenue.”

Creese said that by gaining access to the Cognos customer base, IBM also gains “entree into companies that have already committed to BI. And with IBM using information as a service, they want to own that whole conduit.”

The conduit or “pipe” Creese is talking about is the data pipeline that runs from the database, which IBM owns through DB2, to the transformation phase, which IBM owns through Ascential, to the reporting, which IBM will own through Cognos.

“So the idea is when you go to IBM [or Oracle or Microsoft] you buy a piece of information infrastructure that they’ve connected all together,” Creese said. “You just add the data. You historically had to go to three different vendors for this, but that has changed over time.”

Meanwhile, said Creese, “the part they’re not talking about is that the dance floor is quickly getting empty. After Cognos, SAS is the only big BI player left, but they are privately held. IBM realized if they didn’t buy Cognos they wouldn’t have a whole lot of options.”

Boris Evelson, an analyst with Forrester, said he believes both IBM’s acquisition of Cognos and SAP’s acquisition of Business Objects are defensive moves, “since both companies have been telling us for years that they prefer to grow their BI portfolios organically, with smaller tuck-in acquisition.

“However, organic growth is not happening fast enough, and giving in to sideway pressures from Oracle [with two top of the line BI products from Siebel and Hyperion] and upward pressures from Microsoft [after the Proclarity acquisition and with significant Performance Point market momentum], IBM and SAP had no choice but to react.”

With the proposed acquisition, Cognos customers will benefit from the depth of IBM’s service and support, observers said.

Creese noted that both Business Objects and Cognos have boasted of being agnostic, “but now as part of IBM, will Cognos make sure DB2 is the first version they ship,” he asked. Creese said he did not believe IBM would function that way. IBM’s Mills acknowledged that it would not.

Evelson said many questions arise from the IBM/Cognos deal. “But the main one that puzzles me is, will IBM completely deviate from their strategy of not being in the ‘apps game’ now that it will be competing head to head with Oracle, SAP and Microsoft on BPS [Business Performance Solutions]? What’s next, an ERP acquisition of Lawson or Infor? Or a CRM acquisition of Salesforce.com playing on IBM’s latest Information On Demand strategy?”

Another question is: “Will Cognos be a key component in IBM’s intent to highly optimize enterprises,” Evelson said. “Absolutely yes. IBM is likely to pour significant dollars and resources into making its overall BI [including Cognos] offering ‘process centric’—a key component in our opinion to ultimately optimized enterprises.”

The bad news, Evelson said, is that “all new BI behemoths—IBM, Oracle, SAP, Microsoft—will now be forced to spend more time on product integration, potentially pulling resources away from and reducing priorities of new functionality development.

“Watch remaining smaller pure plays [SAS, Microstrategy, Information Builders, Actuate] and Tier 2 BI vendors [Panorama, QlikTech, Jaspersoft, Pentaho, Logixml, Inforsense and many others] jump on that opportunity. Good news is that the BI market is very hot.”

Tony Baer, an analyst with onStrategies, said: “The other shoe has dropped. But it’s really no surprise at all. Just look at how investors have been bidding up Cognos’ stock since the SAP-Business Objects deal. Given the fact that Cognos is the only Tier 1 independent left, a 9 percent premium on the share price would have otherwise looked rather modest.”

Baer also pointed to a possible Web 2.0 connection.

“The deal does create room, maybe even a vacuum, for midmarket players to fill,” Baer said. “At this point there’s no clear anointed successor to pick up the mantle of BI 2.0: a dynamic, easy-to-use BI approach that’s highly dynamic, borrowing the best of Web 2.0 technologies with innovative approaches to smart data caching. Whoever rises to that challenge should find a very receptive audience.”

Gaurav Verma, product marketing manager at SAS Institute, said: “IBM just now gains business intelligence/query and reporting and financial performance management from Cognos. However, IBM still does not have predictive and advanced analytics or vertical-specific apps.

“SAS is the largest independent software vendor offering companies integrated business intelligence/query and reporting, data integration and analytic capabilities with a long line of vertical-specific applications.”

IBM Slips Further Into Apps Market With $6.8B Cognos DealAbsorbing midtier players lets IBM combine their products with its own in ways that are more seamless than what can be achieved through partnering, says Buell Duncan, general manager for IBM ISV and developer relations. “We can share technology road maps and work more closely day to day to build on each other’s strengths,” Duncan says. “You can do that if you’re one company much more quickly and capably.”

Duncan says IBM will continue to acquire companies whose products are natural extensions of its information management offerings, and whose software–like Cognos’–is built to live in service-oriented architectures. IBM isn’t shopping for a broad-based ERP vendor like SAP or a midtier one like Lawson Software, Duncan insists. “There’s no fundamental shift in our strategy that says we’ll be going into the business applications market,” he says.

Longer term, however, IBM may have little choice but to expand its software footprint to keep pace with major industry players–principally Oracle–that are acquiring more and more pieces of “the stack.” Oracle recently snapped up BI and performance management vendor Hyperion for $3.3 billion, while SAP has agreed to buy Business Objects for $6.8 billion–deals that left Cognos as the only major BI tools vendor without a dance partner.

Still, Duncan insists that IBM is committed to its existing ISV partner program and won’t plunge deeper into applications as a defensive measure. “We’ll continue to work closely with Business Objects” and other partners that have been bought up by rivals, he says.

But there’s another reason IBM may need to build out its app portfolio more aggressively: the growth requirements of a $90 billion company that has become more dependent on software for revenue since shedding big parts of its hardware business in recent years. Duncan is unphased. “We’re not on the same path as Oracle in terms of acquiring pure application vendors for the sake of expanding our customer base,” he says.

Of course, it was only four years ago that Mills told a group of journalists that “IBM is not in the applications business.”

IBM’s planned $5 billion acquisition of Cognos cements the inevitable: the days of buying business intelligence tools from an independent software vendor are drawing to an end. Within one year, the market for traditional, front-end BI software — the reporting and analysis tools used primarily by executives and financial and business analysts and managers — has undergone a major shift from one dominated by independent vendors to one that will soon be dominated by the world’s largest software companies. IBM’s planned acquisition of Cognos follows SAP’s pending acquisition of Business Objects for $6.8 billion and Oracle’s acquisition of Hyperion for $3.3 billion.

With this shift, half of the roughly $5-billion-a-year market for reporting and analysis tools will be owned by SAP, IBM, Microsoft, and Oracle. After that, share sharply drops, with SAS, MicroStrategy, and Information Builders each owning less than 6% of the reporting and analysis tools market, according to an IDC report based on last year’s sales.

The Cognos acquisition will leave privately held SAS as the only remaining independent BI vendor with annual revenues of more than $1 billion (it’s revenues hit $1.9 billion last year), but most of SAS’s sales come from the higher-end niche area of BI known as advanced analytics. Next in size among the independents would be MicroStrategy, which sells primarily reporting and analysis tools and reported revenues of $313.8 million last year.

Both benefits and cautions come with the shift in market ownership of reporting and analysis tools. With the consolidating software industry, CIOs and IT managers regularly cite the benefits of having fewer vendors to deal with and manage. If vendors do the integration work they promise, it might also prove beneficial to purchase front-end BI from the same vendors who sell the apps that contain core business data:Oracle, Microsoft Dynamics and SAP enterprise-resource planning applications, and Oracle, Microsoft and IBM databases.

Yet the remaining independent BI vendors like to argue those large software vendors’ ongoing acquisitions is creating stockpiles of un-integrated software, and could spell trouble for those vendor’s customers.

After SAP’s acquisition of Business Objects, for example, MicroStrategy issued a statement that “Business Objects needed to be acquired” because it was a company weakened by having accumulated so many non-integrated technologies through its own acquisitions, in a market where “organic technical integration” results in the best BI software. MicroStrategy also claimed that unless SAP maintains Business Objects purely as a portfolio investment, it’s likely SAP would change Business Objects’ architecture to improve integration with its ERP software, causing Business Objects customers to potentially undergo “major migrations.”

Cognos Gives IBM The Front-End Tools

IBM’s acquisition of Cognos is an extension of its database-oriented “Information On Demand” strategy it’s talked about over the past few years, which includes many investments in technologies to cleanse, organize and store data. Cognos completes the Information On Demand picture for IBM by providing the front-end tools for accessing cleansed and organized data.

IBM’s Information On Demand has largely been a strategy of acquisitions: data-integration vendor Ascential for $1.1 billion; content management vendor FileNet for $1.6 billion; metadata management company Unicorn; natural language search company iPhrase; customer data integration company DWL; and identity resolution company SRD. Some of that technology, particularly what IBM got from Ascential and Unicorn, make up IBM’s Information Server, a package comprising several WebSphere application servers that conduct various data-integration tasks.

Yet even as they acquire, the big vendors are also making efforts to keep BI choices open to customers. IBM, for example, was likely already engaged in acquisition talks with Cognos when it announced last month a deal it reached with Business Objects, which will soon be owned by SAP and is Cognos’ top competitor. Under that deal, Business Objects will distribute and resell IBM DB2 Warehouse with Business Objects XI and CFO Performance Management software, and IBM will distribute a limited license of Business Objects XI with DB2 databases and warehouses.

Solo BI Vendors Have Something To Offer

With IBM now acquiring Cognos and SAP in the midst of buying Business Objects, the world’s four largest software vendors–IBM, Microsoft, Oracle, and SAP–will soon control more than half of the business intelligence software market. But it’s a mistake to ignore what’s happening in the other half of what IDC reckons is a $7 billion-a-year market for query, reporting, analysis, and advanced analytics software.

Consider part of the reason Business Objects and Cognos are no longer going it alone: Each lost market share last year, according to IDC, and their BI tool sales grew slower than the 12% market average. Most of their revenue comes from tools that run queries against data and generate reports, functions that are becoming commodities. Sales of cheaper Microsoft BI tools, recently rebranded PerformancePoint Server and given an Excel interface, grew 28% last year while gaining a point of market share, to almost 8%, IDC says.

Once IBM completes its $5 billion acquisition of Cognos, announced last week, the only really big BI company left will be SAS Institute, whose revenue will exceed $2 billion this year. Chief marketing officer Jim Davis insists the company–still two-thirds owned by CEO and co-founder Jim Goodnight–isn’t interested in being acquired. It’s among the fastest-growing BI vendors because it dominates the market for advanced analytics, including predictive analysis, and has a loyal customer base. This highly sophisticated (and pricey) form of BI constitutes about 25% of SAS revenue and drives other areas, including data integration software (55% of revenue) and query and reporting tools (10%).

MicroStrategy is the next biggest BI independent, at $313.8 million in revenue last year. CEO Michael Saylor, in a letter to employees last week, predicted that Business Objects and Cognos will veer off on “proprietary strategic trajectories, which place them in a conflict of interest with large segments of their own customer base.” MicroStrategy COO Sanju Bansal predicts the acquired BI vendors will get less cooperation from important allies such as data warehousing vendors. “It’s hard to believe the Teradata folks will open up their technology plan to IBM,” he says, “because the two are mortal enemies.”

But the success of smaller independents such as Information Builders, MicroStrategy, and myriad niche suppliers depends on whether they can provide enough differentiation to be worth the added integration effort. Privately held Information Builders does about $300 million in revenue, but growth sputtered last year as it transitioned from mainframe-oriented BI tools to new markets such as mobile BI access. Revenue is tracking 10% higher this year, says chief strategy officer Michael Corcoran, and its labs are working on user interface technologies that deliver more visual capabilities, including animation, to BI reports. Information Builders also sees opportunity partnering with niche software vendors that compete with the likes of SAP and Oracle–a specialist in ERP for retailers, for example.

Customers will be demanding. Transplace, a provider of supply chain software to the transportation industry, recently bought BI tools from Microsoft, deciding early on that Business Objects and Cognos were too expensive, says Cindy Winkel, director of BI and data warehousing. Microsoft beat out Oracle because it had a package starting with SQL Server 2005 for data warehousing and extending up to OLAP cubes and dashboards, and the ability to view it all in Excel.

Transplace’s decision-making shows the challenge ahead: BI vendors outside the big four still have a lot to offer, but they’ll have to fight that much harder to get considered.

Business Intelligence Consolidation Won’t Kill Innovation, Claim Execs

In a few months, after IBM and SAP will have completed their acquisitions of Cognos and Business Objects, the world’s four largest software vendors will have control of more than half of the $5-billion-a-year business-intelligence software market.

The obvious endgame for IBM, Microsoft, Oracle, and SAP is to sit down with customers and sell them as much of the BI-related software “stack” as they can, extending from databases and warehouses up to the reporting and query tools that sit on users’ desktops. This may appeal to IT managers looking to deal with as few software vendors as possible. But it would be a mistake to ignore what’s happening in the other half of the BI market.

First, consider why Business Objects and Cognos decided to no longer go it alone. Both companies’ market share dropped last year, and both their BI tool revenues grew slower than the market average of 11.6%, according to IDC. Most of their revenues are derived from tools that run queries against data and generate reports, both quickly commoditizing functions in the evolving BI market. Microsoft has been acquiring and improving its own BI desktop tools in the past few years — recently rebranding them all, including Excel, under the name of PerformancePoint Server–and is rapidly gaining share in the BI market. Competition is coming from everywhere: even Google’s enterprise search appliance, in a sense, is a data query tool.

Once Cognos is acquired, the only “large” company left in BI (not including those that sell the deeper data warehouse layer) will be SAS Institute, which will exceed $2 billion in sales this year. Chief marketing officer Jim Davis insisted in an interview Monday that the company isn’t interested in being acquired. Even so, CEO and co-founder Jim Goodnight could easily set his own terms should he ever have a change of heart, and SAS isn’t subject to a hostile takeover: Goodnight, recently cited by Forbes magazine as having a net worth of $8.7 billion, owns two-thirds of the company. SAS is among the fastest-growing BI vendors because it has a near-lock on the market for advanced analytics, including predictive analysis, and a loyal customer base, some of whom have already invested millions on SAS technology. This highly sophisticated (and typically expensive) form of BI makes up about 25% of the company’s revenues and drives other areas, including data integration (55% of revenues), and query and reporting tools (10%). The remainder of the SAS’s business comes from services.

Among the independents, next in size after SAS is MicroStrategy, which reported revenues of $313.8 million last year. In a letter sent internally to staff Tuesday — a copy of which was provided to InformationWeek — CEO Michael Saylor predicted Business Objects and Cognos will veer off ” off on proprietary strategic trajectories, which place them in a conflict of interest with large segments of their own customer base.”

Saylor predicted to his staff that Hyperion and Siebel Analytics development teams will favor Oracle over DB2 or Teradata; Business Objects will be tailored more for SAP apps than Oracle Financials; and Cognos to be “less aggressive” in support of Teradata, Hewlett-Packard, Netezza, Sun, Oracle, and Microsoft.

“Integration and synergy with the parent company will outweigh customer-driven requests for more features, more performance, and more support for existing applications in the current IT production environments,” he added. “We see the chaos and confusion inherent in these mergers as creating a vacuum in the market, and we intend to fill it.”

Microstrategy COO and co-founder Sanju Bansal reiterated the company stance in an interview Monday. “History has shown us CIOs prefer an open stack versus a proprietary stack,” Bansal said. “The second point is there may be conflicts of interest that keep vendors form working together. It’s hard to believe the Teradata folks will open up their technology plan to IBM, because the two are mortal enemies.”

Dave Menninger, VP of marketing at Inforsense, a $7 million-a-year BI company, is among several independent vendor execs claiming that consolidation could actually drive innovation in the BI market.

“If you look at the market as a changing, living organism there’s a process that continually happens,” Menninger said in a Wednesday interview. That starts with small companies that sprout up with good ideas, grow into something successful, often go public or get acquired. But not long after, the best and brightest often split. “When companies get too big, people who are creative and innovative find that constraining,” he said. “They then go to places where they can express their ideas and bring them to market in less than a three-year period.”

Inforsense’s technology, sort of a hybrid of query tools and advanced analytics, is popular among pharmaceutical companies for drug discovery and patient demographics analysis.

Information Builders, a privately held company with about $300 million in annual revenues, also took a positive spin on the consolidating market. “We’re feeling pretty exited about the market, as it seems there will be no slow down in BI as a focus technology in three to five years,” said chief strategy officer Michael Corcoran in a Wednesday interview. “Our focus will be multi-platform-agnostic. As an independent vendor, we have a much better capability to innovate.”

Information Builders’ growth sputtered to a halt in 2006 as the company began transitioning away from mainframe-oriented BI tools to modern-day BI products including more cutting edge areas, such as mobile BI access. The shift is paying off, and revenues are on target to rise 10% this year, Corcoran said. The company is working on user interface technologies in its labs that will deliver more visual capabilities, including animation, to BI reports, he added.

Information Builders also sees an opportunity to build up its channel sales through niche software companies. A developer of customized ERP applications for retail, for example, may prefer to offer its customers Information Builders for analyzing their retail data, rather than SAP-owned Business Objects, noted Corcoran.

The independent BI vendors raise some good arguments. IBM software senior VP Steve Mills, however, claims no such thing will happen with the Cognos acquisition. Just as Cognos will have to continue connecting to non IBM databases, Mills said in a Monday interview, IBM will continue to support other flavors of BI. That, he said, is a given in the “heterogeneous world” of software.

On Cognos:

Cognos Incorporated, a Canadian corporation founded in 1969, is a provider in business intelligence and corporate performance management software solutions. The Company’s solutions helps in improve business performance by enabling planned performance management, supported by effective decision- making at all levels of the organization through the consistent reporting and analysis of data derived from various sources. Its integrated solutions consists of its BI components, performance management solutions, and analytical applications. Its integrated solutions consist of BI components, performance management solutions, and analytical applications. These components are supported by software services for administration, deployment, integration, and extraction, transformation, and loading. In the solution for CPM, it offers products that address the need for organizations to link reporting and analysis to organizational goals and strategies. These solutions span Company functions and processes and define the parameters for performance in scorecards, plans, and budgets. They also help structure, automate, and control processes that relate to governance and compliance such as consolidation and financial reporting. Its software services consist of the Administration Services, Deployment Services, Integration Services, ETL Services.The Company’s worldwide sales and marketing organizations are led from Burlington, Massachusetts location. The Company has global customer base with over 23,000 customers located in more than 135 countries. Its primary target market is Global 3500 companies and large public sector organizations. Cognos, the Cognos logo, Axiant, Cognos DecisionStream, Impromptu, NoticeCast, PowerHouse, PowerPlay, and Cognos ReportNet are trademarks or registered trademarks of Cognos Incorporated in the United States or elsewhere. As of February 28, 2006, the Company had 3,516 full-time permanent employees.

More on Cognos

Cognos to Acquire Applix for $339 Million:

Applix, of Westborough, Mass., will complement Cognos’ latest BI software—Cognos 8 Planning, 8 Controller and 8 Business Intelligence—by bringing additional capabilities for analysis and optimization of complex financial performance management, officials said. Applix will bring finance self-service modules with a business rules engine, as well as Applix TM1, the company’s 64-bit in-memory multidimensional OLAP (online analytical processing) server.

“This acquisition is a terrific strategic fit for Cognos. Applix will broaden our solution offering and provide Cognos with an innovative, 64-bit, in-memory analytics capability,” Cognos CEO Rob Ashe said in a statement. “It will also bring into the company a very strong employee and customer base that has been committed to performance management through high-impact analytics. This is another major step forward for Cognos in delivering leading performance management solutions to finance and across operations.”

Cognos, IBM Team Up on Bank Risk Management

The Cognos-IBM Risk Adjusted Profitability Blueprint, announced Aug. 22, is geared toward helping companies—mainly banks—better integrate risk information with revenue and expense planning data.

The goal: to help banks maximize risk-adjusted return on capital, striking a better balance between aggressively expanding their businesses and controlling risk exposure.

Risk factors for banks include risk-adjusted return on capital and economic capital, probability of default, loss given default, and expected loss.

The co-developed blueprint combines Cognos 8 Planning and Cognos 8 Business Intelligence with IBM Banking Data Warehouse. It also brings in financial services expertise garnered through IBM Global Business Services.

Overall, the combined reporting environment includes functionality for integrated risk-adjusted planning and forecasting across multiple dimensions with built-in risk factors. Also included are activity-based management and costing for customer, product and business segments, which provide profitability planning and reporting data.

A multidimensional scenario-planning module incorporates activity costs—data that allows users to compare the financial impact of several business options and take action.

The Profitability Blueprint, available now, is not the first bank-related Blueprint Cognos has developed. The company also developed blueprints for branch performance, corporate and retail banking, customer segment performance, and insurance product profitability

Cognos Now Is Newest Member of Cognos BI Family

Cognos announced May 14 a new family of appliance-based performance management and business intelligence software. Cognos Now also is available as a service.

The appliance consists of operational BI—a sort of generalized roll out of BI and analytic capabilities designed for the everyday user. Users are able to create self-service dashboards, analyze data and write reports with minimal to no IT input. The software also includes in memory 64-bit computing capabilities culled from Celequest’s patent stable, and data integration or ETL (Extract, Transfer, Load) functionality.

On the hardware side of the Cognos Now appliance, the company has bundled a rack-mounted server with a J2EE open-source application server and a meta-data repository.

Cognos will offer three configurations: one for the lower end of the market, a standard edition and an enterprise edition. What varies among the three editions is the amount of CPUs utilized and the amount of memory installed. The enterprise edition also has some capabilities for integrating to other applications such as SAP’s Enterprise Resource Planning transactional applications.

The company also will offer different variations of the appliance.

The first, available now, is Cognos Now for Salesforce, geared towards customers who need to monitor performance against their Salesforce.com data.

The idea with an appliance-based BI is to enable companies or divisions without a lot of IT support to create their own dashboards, modifications and watch points without having to involve IT, according to Jim Hare, vice president of Product Marketing and Business Development, Cognos On Demand.

“The idea is to enable [users] to make changes without having to go to IT,” Hare said.

Both the appliance and SaaS capabilities are lifted almost entirely from Celequest, a company Cognos acquired in January. Celequest had developed a BI appliance called Lava that included ETL, streaming OLAP analytics, self-service visual dashboards and a built-in rules engine for event-driven analysis. On the hardware side, Lava took advantage of multi-core processes for the standard editions. The higher-end enterprise edition included two multi-core CPUs, up to 16 gigabytes of memory, hard disk drives and was pre-wired for network infrastructures.

Cognos has tweaked Lava to include newer Web 2.0 concepts such as Google maps. It’s also interoperable with Cognos 8 Business Intelligence, which provides a more comprehensive BI platform.

Like Lava before it, Cognos Now is available as an appliance or through the Software as a Service model. The appliance (and SaaS) model represents a cheaper, albeit scaled down BI tool for some users. However, some critics warn that a pre-configured appliance can only be used for its specific purpose; it can’t run any other applications in the same box.

Nonetheless, Cognos is not alone in its pursuit of appliances that pre-configure software and hardware for a specific need. SAP announced its BI Accelerator last year and it’s largely anticipated that its new A1S mid-market suite, in the works now, will be available on demand, with some functionality pre-configured and bundled on an appliance.

Earlier this month, BI software vendors MicroStrategy and SaS confirmed that their BI software is being tailored for HP’s NewView appliance, according to media reports. In March, Business Objects unveiled its Open Appliance Initiative that amounts to partnerships with data warehouse vendors—IBM, Netezza, DatAllegro, Greenplum, HP, RPath, Teradata (a division of NCR) and VMware—which will pre-install Business Object’s BI software on its hardware to create mid-market geared appliances.

Cognos Acquires Celequest to Speed BI:

By adding privately held Celequest to its roster, Cognos will achieve several objectives: a longer reach into channel markets, a better breadth of offerings for midmarket and on-demand customers and, at the end of the day, faster business intelligence capabilities.

Celequest, based in Redwood City, Calif., develops a BI appliance that utilizes in-memory, or streaming data, technology that stores data in memory rather than at the database level. The result is much faster query times on BI requests, with the capability to support thousands of concurrent users with little latency of information. Since the capability, dubbed Lava, comes in the form of an appliance—essentially hardware that plugs into a network—users are able to deploy BI functionality in a short time frame (days, according to Celequest’s Web site).

There are three versions of the Lava appliance: an SMB (small and midsize business) edition, a standard edition and an enterprise edition. Each includes performance dashboards, an application workbench to set up data integration and an analytics server that houses the in-memory technology.

Cognos Search Tool to Support New Enterprise Software Offering from IBM and Yahoo

Cognos, Panorama Launch New BI Platforms for SAP

Cognos Gives Banking Industry BI Blueprint

Cognos to Add Search to BI Platform

IBM, Cognos Look to Ease SOA Deployments

Cognos and FileNet Agree to Integrate Products

Running a Hatchery for Replicant Hackers

October 22, 2007


At 26, Phil Yuen’s identity lay somewhere between Dilbert and a Microserf. His office, on the first floor of Building 16 on the Microsoft campus in Redmond, Wash., did not have a window. Not to worry. “I could see the window from my manager’s office,” he said.

Mr. Yuen was a midlevel manager who wrote specifications for enterprise project management software — giant computerized flow charts. “I drew boxes and lines to help other people draw boxes and lines,” he said.

Then one night last spring, he was sitting in his apartment, surfing the Web. He went to Slashdot.org (“news for nerds”) and saw an article that Paul Graham, the essayist and guiding light of computer programmers, was establishing a start-up company, later called Y Combinator, which would be giving seed money to hackers to start businesses. The word “hacker” is not derogatory in the computer world, but it is someone who is creative and resourceful with code.

“It’s like Rob De Niro wants to start an acting school,” Mr. Yuen said. “Do you want to join it? You get to work with him every week, you might even get a small little movie deal out of it at the end.”

Mr. Graham, a 41-year-old dimple-cheeked entrepreneur who sold his company, Viaweb, to Yahoo in 1998, had developed a large following for his lucid and contrary essays in a geek community more comfortable expressing itself through programming code than coherent paragraphs. Among his essays, “How to Start a Startup” is a siren call, resonating with Gen Y programmers pondering quarter-life crises and with college students too inexperienced to be jaded.

After the establishment of Y Combinator, more than 200 teams, with two to four people a team, applied in just a few days for eight slots announced by the new company, which operates in Mountain View, Calif., and Cambridge, Mass.

After being accepted, Mr. Yuen and his team established two start-ups in the last seven months: one he gave up on; the other, called TextPayMe, is a service that sends cash payments to an online account through text messaging — akin to PayPal but using a cellphone.

Last month, Mr. Yuen was in Mountain View, in the heart of Silicon Valley, demonstrating his product to venture capitalists and other tech entrepreneurs. At a Y Combinator dinner, he asked Evan Williams, a co-founder of Blogger, who was a guest speaker at the event, for his cellphone number and then sent him $2 using TextPayMe. (Mr. Yuen, incidentally, has now accumulated a trove of cellphone numbers of Silicon Valley heavyweights and offers from three venture capital firms.) Mr. Williams, intrigued, mentioned he had the same idea earlier that day while brainstorming with friends. “It’s something in the air,” he said.

That night, at the dinner, Mr. Williams addressed a room full of refugees from Oracle, AOL, Microsoft and Accenture. “I think this is so cool what you guys are doing,” said Mr. Williams, who sold his company to Google in 2003. “I would love to be part of something like this. I’m so jealous.”

If it is possible to systematize the archetypal two guys in a garage (and they are generally guys), the year-old Y Combinator wants to do it. The company’s formula is to throw smart people together and provide them $6,000 in seed money per person to cover the initial costs of the company, cookie-cutter legal paperwork and an extensive network of business contacts.

In return, Mr. Graham and his partners — Jessica Livingston, Trevor Blackwell and Robert Morris — collectively own 1.2 to 12 percent of the company, with an average of 6 percent. The company holds two boot camps a year for about eight groups each session, a summer one in Cambridge and a winter one in Mountain View. Y Combinator is not so much an incubator as a hatchery for baby companies, and as with all things spawned in bulk, some will die, some will flourish and some will eke by.

“Y Combinator comes down to two kids in a room with two computers and ramen noodles for a summer,” said Chris Sacca, a principal of new business development at Google and a speaker at Y Combinator’s one-day start-up school conference in October at Harvard. “It takes ambitious geeks and puts them in a situation with no distraction and expects audacious outcomes from them. The reason we like it is that that is what Google is.” Indeed, Google has made acquisition overtures to one of the companies that was formed during the summer session, which the founders turned down.

Mr. Graham got the idea for starting Y Combinator after giving a talk to student entrepreneurs at Harvard, where he received his Ph.D. in computer science. He told them to look for seed money from rich people they knew, preferably ones who had made their wealth from technology. “Then I said, ‘Not me,’ and they all looked kind of downcast and then I felt like a jerk,” he said. Then, on reflection, Mr. Graham thought, why shouldn’t he try to support young hackers?

The goal for Y Combinator’s young entrepreneurs is twofold: to make something people want (which is the company’s motto); and to stretch their financing long enough for additional investment or to get acquired. For instance, Mr. Graham’s former company, Viaweb, which made software to build commercial Web sites, was bought by Yahoo and reborn as store.yahoo .com. One Y Combinator business created last summer, a company that uses cellphones for social networking, got financing from a venture capital firm. Two other Y Combinator companies, a calendar Web site called Kiko.com and a news site, Reddit.com, received additional angel financing.

Y Combinator relies on certain premises: that open-source software and falling hardware prices means that tech start-ups are cheap to finance; that large companies are no longer at the forefront of innovation; and that mature technology companies find it cheaper to buy than to build.

The company takes its name from an obscure mathematical term, describing a function that generates other functions. Y Combinator is a company that creates other companies — a sly reference that would elicit a smile from a very narrow set of people, but luckily the same set that the company is trying to appeal to. It is the philosophical triumph of the passionate computer hacker over the uptight M.B.A.

“Paul is telling us, ‘If you are having a good time, and you are building something that other people want, then the money is not going to be a problem,’ ” said Beau Hartshorne, 24, whose vision for a company he is calling Pixoh will allow people to resize and crop their photos within a Web browser.

Mr. Graham is more focused on creating cool products — that is, coding as art — than developing revenue models and protecting intellectual property. Thus, Y Combinator may not be as good at teaching participants how to build self-sustaining companies than it is preparing them to sell, or flip, their businesses. For Silicon Valley corporate war chests, acquisitions are often made for technical talent as well as product, which generally has to be rebuilt if it is kept at all. The whispered acquisition rate for companies is about $1 million to $2 million per technical employee.

“It’s a way for recruiting for Yahoo where you don’t have the risk and uncertainty of knowing if they can actually do something,” said Joel Spolsky, a technology entrepreneur in New York City who also has a wide readership on the Internet.

“The danger is that you will run out of money before a buyer shows up,” said Peter Rip, a managing director of Leapfrog Ventures, a venture capital fund in Menlo Park, Calif. “We are seeing a lot of guys being attracted to the ease with which they can start a company, but starting is really only the first step.”

From a financial standpoint, the Y Combinator investments are small in a world where valuations are measured in millions of dollars, if not billions. “In the traditional venture capital model, $6,000 isn’t enough to pay your corporate lawyer fees for your first financing,” said Mr. Sacca of Google.

“It’s not just about the money,” said Jeff Mellen, 24, who quit Oracle with three of his friends to build an operating system that works within a Web browser.

Y Combinator provides the validation that young techies should keep pursuing their dreams. “I could get the same amount of money from my parents, but that wouldn’t tell me if my business plan or idea was a good one,” Mr. Mellen said. “That tells me my parents love me.”

Source: New York Times

Y Combinator Inspires Imitators

September 26, 2007

by Kerry Miller

Paul Graham knows that most startups will eventually fail. But at Y Combinator, the hacker guru’s seed fund for young techies, they’re encouraged to fail fast and fail cheap—and then to reboot and start again. Designed with the new economics of Web-based entrepreneurship in mind, Y Combinator is a new type of venture fund—dispensing tiny amounts of cash but lots of hands-on mentoring.

Since its founding in 2005, Y Combinator has funded 58 startups—nearly all in software or Web services—bringing in a new crop twice a year (in Cambridge, Mass., in the summer; in the San Francisco Bay Area in the winter) for a three-month crash course in the Graham-ian philosophy of entrepreneurship (BusinessWeek, 9/26/07). Each team gets a small sum to cover basic expenses ($5,000, plus $5,000 per founder) and the chance to pitch their idea to a roomful of top angel investors and venture capitalists. The insider status of Graham himself—a Web software pioneer who sold his startup to Yahoo! (YHOO) in 1999 for $49 million—lends additional cachet.

“Cheaper Equity”

In return, Graham and his four partners get a stake in the company, usually about 6%. So far, Y Combinator investments like the social news site Reddit, acquired by Condé Nast in 2006, and slide show app-maker Zenter, acquired by Google (GOOG) earlier this summer, both for undisclosed amounts, are positive signs that the formula works. Because the sums Y Combinator invests are so small, even early-stage acquisitions can translate into relatively big paydays.

After two years, Y Combinator’s portfolio is still too young to fully evaluate the success of Graham’s strategy—but that hasn’t discouraged a growing number of investors around the globe from launching their own copycat programs.

One main reason? “It’s cheaper equity,” says David Cohen, the executive director of the most established Y Combinator counterpart, a Boulder (Colo.) program called TechStars, which ran its first three-month session in May, 2007. “If you look at the companies that are getting to an A-round, the model that we’ve created gives a significant discount to those valuations.” Critics say that makes the arrangement exploitative for the startups involved, but Cohen disagrees: “I think for a first-time entrepreneur, it’s actually an incredible deal.”

Both Cohen and Graham say money is actually among the least important things they offer to entrepreneurs, since drastically cheaper hardware, software, and bandwidth—among other factors—have made it much easier to start a company than it was a decade ago (BusinessWeek, 5/31/07).

Dating Before Marriage

But disappearing barriers to entry also make it harder for any one startup to stand out in the crowd—a challenge that’s as tricky for investors as it is for entrepreneurs themselves. Without a track record to go on, after all, how do you discern a future Mark Zuckerberg (Facebook) or Kevin Rose (Digg) from a sea of equally fresh-faced, Red Bull-swilling computer whizzes? With a program like Y Combinator, investors get three months of hands-on experience with talented entrepreneurs—and an early look at their work—before reaching for their checkbooks.

Cohen says he’s already fielded inquiries from angel investors in 20-odd cities, all interested in starting similar programs of their own. “You’re going to see a massive proliferation of this model,” he predicts.

Graham, a widely read essayist who has written prolifically on topics like “What Business Can Learn From Open Source,” isn’t flattered by the unauthorized open-sourcing of his own model—including one clone in Vienna shamelessly named YEurope. (His opinion about such copycats is made clear on Y Combinator’s FAQ. Question: “Will you help us set up something like Y Combinator in our town?” The answer: “There already is a Y Combinator in your town: Y Combinator.”)

Why Reinvent the Wheel?

Many of Graham’s admirers, however, say they respectfully disagree. Among them is Web industry veteran Saul Klein (BusinessWeek, 6/7/07), a partner at London venture capital firm Index Ventures and a former executive at Skype (EBAY). Klein helms the most prominent Y Combinator lookalike in Europe thus far: the London-based Seedcamp, which launched in September. In a slightly novel twist, Seedcamp began its competition by bringing 20 startups—plucked from an application pool of 268—for an “unconference” (BusinessWeek, 5/14/07) before selecting six companies to participate in the three-month program. Each of the six companies received €50,000 (about $70,614) in seed money—a considerably heftier sum than other Y Combinator-like programs—with Seedcamp taking a larger, 10%, stake in each.

And bona fide venture hubs like London aren’t the only place the Y Combinator model is attracting interest. Cities such as Lexington, Ky., and Milwaukee are taking note, too. In Atlanta, a small group of angel investors hopes to replicate the Y Combinator model with a program called BoostPhase. Co-founder Wayt King says he envisions BoostPhase, set to launch this fall, as “a Y Combinator for the Southeast.”

“We have huge respect for Paul Graham and what he’s done, and we figured there’s not much sense in reinventing the wheel,” King says—adding that he doesn’t see BoostPhase as competition for Graham’s original. “There are a lot of entrepreneurs who don’t want to or simply can’t move to Silicon Valley or Boston.”

Standing Out from the Crowd

TechStars’ Cohen says he’s not worried about the competition. “There’s a pretty clear demand for this,” Cohen says, noting that the 10 startups TechStars funded represented only 26 of 302 applicants. Y Combinator accepted an even smaller number for its summer 2007 round—19 of 435 applicants—making an applicant’s chances of getting tapped for Y Combinator (4.4%) only slightly higher than those of the brainiacs vying for a Rhodes scholarship (last year’s acceptance rate: 3.6%).

For participants, the real draw of a program like Y Combinator isn’t the money, it’s credibility—and not just with investors. The traffic bump from an early mention about getting funded on an influential blog like TechCrunch can be just as crucial as a Series-A round in helping a Web startup gain traction over its rivals. And by adding structure to an otherwise nebulous pursuit, Y Combinator also makes a startup a less risky endeavor for entrepreneurs themselves—both practically and psychologically. “Y Combinator provided a socially acceptable way to drop out or postpone my college education and focus on the company full-time,” says Kevin Fischer, a 21-year-old industrial engineering student at the University of Pittsburgh who applied to both Y Combinator and TechStars. Plus, he adds, “It seems like even the people that fail still go on to jobs at Google—no one does too badly.”