Don't expect IBM to become a champion of SAAS any time soon.

After a visit to IBM's Somers, N.Y., campus, I am convinced that while IBM's heart is with the Web, its feet are still mired in 20th century technologies, putting it at odds with history and the best interests of its customers.

It wants to sell SAAS but is hamstrung by a compensation system that only rewards sales of large on-premises contracts. And while it uses collaborative tools internally, it tethers the Web 2.0 products it brings to market to legacy software systems that are rapidly becoming obsolete.

IBM currently supports partners who want to offer their customers on-demand applications with cloud-based infrastructure like the iDataplex server.

It also continues to support the cloud offerings of companies it has acquired, such as cloud storage from Arsenal Digital Solutions or business intelligence reporting from Cognos.

But IBM is still approaching SAAS very timidly.

As Clint Boulton wrote about the iDataplex launch:

IBM is picking its spots in the industry, for good reason. Its core business segments are servers, software and global services. If the company took its entire software group and hosted it in the cloud for customers--a la Google or Yahoo--it would begin to eradicate its server and global services businesses.

Bluehouse, part of the Lotus family of collaboration tools, is probably the most significant native IBM SAAS offering to speak of, agreed Katharine Frase, IBM's vice president of technical and business strategy.

I asked her why IBM demonstrates so little enthusiasm for SAAS.

She ticked off a list of drawbacks to SAAS straight from the on-premises software vendor playbook dated 2005: the data might not be secure; SAAS won't succeed with large companies; you can't customize it; integration becomes the customer's problem.

Try telling that to Marc Benioff, who with Salesforce.com has proven that big companies want SAAS, that it can be customized, and that it's secure enough that large financial services firms are not only using it--they're among his biggest customers.

When I told her that I found her reticence archaic, Frase admitted that, actually, IBM is still struggling with internal issues, such as how to change the compensation structure based on commissions generated by selling on-premises software.

This kind of issue is "internal machinery that doesn't matter to anyone else," she said. "What you'll see over time is less reticence."

She also noted that IBM is "seldom the very first in adopting a new trend, but we catch up quickly."

OK, but in the meantime, customers will either have to wait or find another vendor who gets it now.

The irony is that, while Frase is fighting organizational entropy in Building 1 of the Somers complex, Linda Sanford, IBM's senior vice president of enterprise on-demand transformation and IT, is across the campus in Building 3, trying to help customers become globally integrated enterprises, in part by showing the way herself with IBM as exhibit A.

IBM is its own walking cautionary tale; many people there still have vivid memories of staring into the abyss; they understand that paradigms can shift in the blink of an eye.

In fairness, IBM tries to practice what it preaches and uses crowd-sourcing technology to capture the wisdom of its masses (more than 500,000 strong worldwide) in virtual meetings it calls Jam sessions.

The company also boasts of more than 26,000 users of Beehive, IBM's internal social network, and more than 15,000 internal wikis accounting for over 250,000 pages and more than 214,000 users.

IBM also claims to have more than 155,000 members in its LinkedIn user group and 24,000-plus members on Facebook.

Sanford said that internal "technical adoption programs" led to the development of technologies that Big Blue brought to market as Sametime 7.5. But Sametime 7.5 was launched in 2006--well after the Web 2.0 revolution had gotten under way.

That timeline doesn't speak highly of IBM's capacity for innovation.

Moreover, IBM wants to provide customers with Web 2.0 tools like IM and social networking through existing proprietary technology anchored to its legacy Lotus application.

But customers--actual knowledge workers--want these technologies free of encumbrances and, whenever possible, free of charge as well. There isn't anything available from IBM that isn't available for download online from someone else with a more modern conception of Web 2.0.

IBM would argue that those tools are missing security and compliance features that enterprises need. But those issues have never stopped corporate adoption of consumer tools before and won't in the future.

Despite Sanford's efforts, the old IBM culture isn't going away quietly. Sanford didn't disagree when I told her that customers still complain about IBM salespeople giving the same laborious presentations as ever and dragging their feet when asked to interoperate with competing products.

"This is a journey. Transformation doesn't happen overnight," she said.

Indeed, you can't change culture just by saying you want to.
As Tom Davenport, a noted expert in organizational dynamics, wrote:

Enterprise 2.0 software and the Internet won't make organizational hierarchy and politics go away. They won't make the ideas of the front-line worker in corporations as influential as those of the CEO. Most of the barriers that prevent knowledge from flowing freely in organizations--power differentials, lack of trust, missing incentives, unsupportive cultures, and the general busyness of employees today--won't be addressed or substantially changed by technology alone.

Sanford and Frase think IBM can pull off the transformation. I have my doubts. I'm not sure that you can pretend to be one thing when you're built off a different blueprint. I don't think you can take a baseball team full of sluggers and turn them loose on the bases. Or to use a more international sports analogy, I don't think you can take a soccer team built to score off set pieces like corner kicks and transform them into a team that scores by rapid ball movement.

IBM's customer-facing façade has too many holes for its organizational spackle to successfully hide. The holes include a tepid commitment to SAAS and a hidebound approach to Web 2.0 technologies.

It's trying to please too many constituents--partners who don't want IBM to sell applications that compete with their own enterprise customers who don't trust their employees with blogs and wikis, and its own sales force, which is married to a compensation structure that is ill-suited to selling modern tools in the enterprise.

Customers looking for Web-based tools and software delivery mechanisms can wait around until IBM figures out how to deliver what they need--while still meeting the growth targets that Wall Street expects. Or they can look elsewhere.