, the scrappy hosted CRM provider that calls for “the end of software,” found its own software at an end yesterday.
While some competitors rushed to cast doubts on the company’s offering — and on software-as-a-service in general — others pointed out that, well, stuff happens.
Salesforce.com confirmed that on Tuesday, some salesforce.com users experienced intermittent access from approximately 9:30 a.m. to 12:41 p.m., and then again from around 2:00 p.m. to 4:45 p.m. EST.
The company blamed the outage on an error in the database cluster on one of the company’s four global nodes; the database is provided by an outside vendor.
In an e-mailed statement, a company spokesman wrote, “All four global nodes are currently operational and running normally. There are no outstanding issues in the system at this time. No other aspects of the system were involved.
The company claims around 350,000 users of its basic customer relationship management offering, as well as a multitude of custom applications built on its sforce platform.
Salesforce.com didn’t respond to a question about whether or how it communicated with customers about the outage. Jonathan Tang, CEO of salesnet, a competitor, said that while unscheduled down-time is unavoidable, companies should alert customers immediately when there’s an outage and keep in touch with status reports.
Salesnet has four tiers of customers; those at the top can expect hourly calls from account executives and engineers during a “code red,” while the lower tiers can expect e-mails to their administrators.
Tang acknowledged that there’s growing concern among customers about the scalability stability and performance of software-as-a-service (SaaS). “Companies don’t have control over the network and data center,” he said.
Tang said salesnet has “warm” backup centers that replicate the entire application: servers, software and database. The entire operation is backed up every 15 minutes.
“If we think there could be an outage, we do a warm failover,” he said.
Analysts have been bullish about software on-demand. IT research firm Saugatuck Technology sees increasing adoption of both software-as-service and what it calls “pay-as-you-go” models for both core infrastructure and business applications over the next five years.
In a November research note, the company said that these models are becoming popular among enterprises because they tie the cost of the software to the value it provides to the user.
Still, the application categories that the firm said were ripe for adoption of the on-demand model are not mission-critical. According to Saugatuck, the top four are travel services, benefits administration, payroll and billing.
Arnie Berman, chief technology strategist at SG Cowen, recently told internetnews.com, “I think you’ll find [software as a service] successful in niches where companies have specific needs and uses and can save money.” For example, he said, if a hosted expense report application doesn’t work, it’s no big deal.
Saugatuck research analyst Bruce Guptill said that, while repeated outages would be a serious problem, this was the first since some brief salesforce.com downtime a year ago.
“To say that businesses should not rely upon SaaS or other on-demand applications and services because of potential outages is disingenuous at best,” he told internetnews.com. “One may as well claim that network or telecom outages are a reason not to use those services for mission-critical business.”
Mark Murphy, managing director of investment bank First Albany Capital, said that while competitors might try to use yesterday’s bummer to their advantage, he’s highly optimistic on the future growth potential for Salesforce.com and the on-demand sector.
Murphy retained his neutral rating on the company in the face of the news, and he expects earnings per share of CRM for fiscal year 2007 to be $0.24.
SaaS vendors defended their model, emphasizing that all applications go down.
“The on demand delivery model is more than capable of handling mission critical functions,” said Greg Gianforte, CEO of RightNow Technologies, a provider of a multi-tenant hosting environment for CRM applications. He noted that more than half of the company’s revenue last quarter came from organizations with more than $1 billion in revenue, including major government agencies and educational institutions.
Simon Peel, vice president of technology for Cast Iron Systems, a provider of on-demand application integration, said, “Saying you shouldn’t rely on on-demand computing because there are occasional issues is like saying you should build your own electricity plant because the power might go out once a year.” Cast Iron is a salesforce.com partner.
According to Peel, the issue is the reliability of the particular application, not of the on-demand model. “When it comes down to it, I’d rather pay $65 a user and get an army of IT people at Salesforce.com fixing issues as they come up, [rather] than install Siebel locally and have to pay for that army of IT people myself.”
Acknowledging that nearly six hours is a painful outage for anyone, Treb Ryan, CEO of OpSource, said, “Salesforce.com’s uptime is way better than any corporate application I know of. How often are even simple apps, such as e-mail or the phone, down in most organizations?”
OpSource provides infrastructure and support for software vendors that want to offer their applications on demand. Ryan said that the general expectation is for much better uptime than what salesforce.com experienced on Tuesday. But he pointed out that complex on-premise applications seldom run with anything like the uptime of Salesforce.com. “Every company I’ve ever worked for was constantly having the ERP and CRM and HR apps go down for maintenance, network problems or general outages,” he said.
Ryan had one final thought: “When Salesforce.com is down, they don’t get paid,” he said. “When one of my regular enterprise apps is down, I’m still out of the money I paid for it.”