Visitor Monitoring Forecast: Partly Cloudy and a Chance of Applications | Keynote

Forecast: Partly Cloudy and a Chance of Applications

The promise and performance challenges of Software-as-a-Service

upgrade to their core enterprise application offering went flawlessly. Monthly subscription revenue is up by high double-digits. Customers are happy. The press veritably glows about their cloud-based application’s wonders. And Google, Microsoft, and a host of VCs and investment bankers are lighting up the phone lines with offers of truckloads of cash.

SaaS At Last?

After years of latency, software-as-a-service is moving to center stage, swept along with the bigger concept of cloud computing. On the software side, applications like Salesforce, Workday and Freshbooks — and yes, GoogleApps — have become serious players in enterprises of all sizes. And in the bigger cloud, services like Amazon’s EC2 Elastic Compute Cloud and IBM Smart cloud services — and yes, Google AppEngine — are changing the way IT departments approach their missions and their development tasks, enabling greater speed and flexibility than ever before.

Forrester Research counts “ubiquitous deployment of software-as-a-service for packaged applications” as one of its 15 tech trends for the next three years. 1Forrester Blogs, “CIOs: Develop A Technology Watch List,” by Sharyn Leaver, 11/6/09 Gartner predicts a 23.8 percent compound annual growth rate for SaaS through 2012, more than doubling overall market growth of 11.4 percent. 2Gartner, “Market Trends: Software as a Service, Worldwide, 2007-2012,” Sharon A. Mertz and others, 9/12/08 And according to a Piper Jaffray report quoted in Information Week, spending on the cloud will outstrip the overall software market by a factor of five. 3Information Week/Plug Into the Cloud blog, “Piper Jaffray Sees Gold Rush In Cloud Software,” by Charles Babcock, 2/23/10

We’ve seen this type of technology trajectory before and can recognize its course; think back to the Web itself, to early social networking, to the still-snowballing mobile data market. The dominant SaaS providers are still rising to the top; economics, standards and best practices are being sorted out; business models are being built and business cases made. The difference is that these sweeping technological changes keep happening faster and faster.

Many point to as a success story and example of what’s possible. Indeed, a Gartner consultant recently put Salesforce’s share of the total SaaS spend at 14.6 percent. 4Gartner webinar, “Ensure Your SaaS, Cloud Investment Will Deliver Results and Save You Money,” 4/2010 But without certainty as to the potential, nor enough track record to know and avoid the pitfalls, how willing will enterprises be to trust their critical applications to cloud-based solutions that ultimately are out of their control? A few stories in the press about outages at Google, Facebook, and Twitter are enough to give any business decision-maker pause. Even so, the SaaS train has apparently left the station. The question is, at what stop will businesses get on, and for how long a ride?

Cloudy Terminology

To the average business person, the whole concept can be a little — cloudy. In the most general sense, “the cloud” refers to IT infrastructure and functionality that exists outside a company’s walls (except in the case of “internal clouds,” but that’s another story), whether it’s for data infrastructure (Infrastructure as a Service), processing power (Platform as a Service), or business applications (SaaS). These things collectively make up “the cloud.”

While all aspects of the cloud can and do impact businesses as a whole, as well as individual users, it is software-as-a-service that is the most visible and has the most direct effect on how people do their jobs every day. SaaS either replaces or supplements the on-premises or desktop applications that workers use day-in and day-out, whether it’s simple word processing or complex CRM or HR functions. In some cases, most notably Microsoft and its Office suite, hybrid solutions combine familiar desktop applications with added functionality from the cloud, such as enhanced collaboration, online file storage and anywhere-accessibility.

Such hybrid configurations may serve as the gateway to cloud applications for many companies. But by any route, it is likely that SaaS is going to become business as usual. Even Bill Gates saw it coming while still running Microsoft, describing the “services wave” as “the next sea change. 5Trumba, “White Paper: Five Benefits of Software as a Service,” 3/8/07

Anshu Agarwal, vice president of marketing for Keynote Systems, sees users themselves as important drivers of SaaS adoption. “What I have seen is that, after years of being dependent upon internal business system schemes, users and companies have become tired of having to deal with upgrades, installations, and internal bottlenecks,” he says. “Software-as-a-service is being driven as much by internal users as it is by software-as-a-service companies. It comes down to users saying ‘I’m just tired of having to wait for what I need. I can turn this on next week. I don’t have to wait in a queue, I don’t have to be denied the kinds of automations I want.’ I think that’s what’s driving it.”

Promises, Promises

It’s a great idea on paper — replacing big outlays for software and maintenance with flat monthly fees, fast implementation, never having to worry about upgrades, slashing hardware expenses — but the SaaS cloud may have to come closer down to earth for these promises to be realized. Some of them are more in reach than others.

Lower Cost. It’s a reasonable conclusion that initial implementation cost can be significant lower with SaaS than with on-premises software. Forrester Research says SaaS implementation can typically run .5x to 1x the first year’s subscription fee, vs. 1x to 5x for traditional software. 6Forrester Research, “The ROI Of Software-As-A-Service,” by Liz Herbert and Jon Erickson, 7/13/09 Total cost of ownership, however, is not so clear-cut. SaaS TCO can be lower if it brings corresponding reductions in IT hiring, hardware purchases, and database and app server licenses. 7Gartner webinar, “Ensure Your SaaS, Cloud Investment Will Deliver Results and Save You Money,” 4/2010 The balancing of price and performance will ultimately determine whether money is saved, or value added for a comparable investment. One big advantage SaaS does have is that costs are knowable, and predictable, based on fixed monthly per-user fees.

Rapid Deployment. Speed of implementation is a real advantage for SaaS. Companies have no need to gear up with hardware or devote IT resources to installation, testing, and rollout. All that is done on the vendor side. The trade-off, though, is that implementations are fairly standard, with limited opportunity for customization. If it fits as is, great. If not, business processes need to be adjusted to accommodate the software, instead of vice versa. For some, that is a serious trade-off.

Seamless Upgrades and Maintenance. Another checkmark in the plus column for SaaS — instead of the highly disruptive and expensive process of upgrading on-premises software, SaaS upgrades happen on the vendor side, often without any interruption of normal business processes. And with SaaS, upgrades and fixes can happen more frequently. The SaaS vendor team is dedicated to one system that serves many clients — with no string of legacy products to support — and therefore can bring innovation more quickly to client companies with none of the stress and turmoil of traditional upgrades. As an example, in the ERP arena, in the five-year period from 2004 to 2009, SAP rolled out just two product releases, and Oracle and PeopleSoft three each. During the same period, Workday, a SaaS competitor, implemented nine product releases. 8Knowledge Infusion Center of Excellence Research, “The Continuous Innovation Advantage of Software-As-A-Service,” by Jason Corsello, 10/2009

And underlying all these benefits is the potential for a now-smaller corporate IT staff to focus on gaining competitive advantage instead of routine housekeeping and utility work.

The Elephant In The Cloud: Performance

For all of its tremendous potential, there’s one thing that will make or break the SaaS model both for vendors and users: Performance. No longer is technology contained within the walls of the enterprise, running on its own proven network, controlled closely by its own IT department. Now, the nerve center of the enterprise, the productivity of workers, the integrity of information assets is controlled by an outside entity, flows through the various pipes from a remote data center over the Internet into the enterprise’s network and ultimately, into a browser. And that presents challenges both for performance and user experience.

“My browser is not like a dedicated desktop application,” Agarwal says. “Take for example, Outlook. It’s fired up in the morning and it synchronizes and voilà! The mail is there. But I can also use Web mail, right? And it’s a very different experience.”

“It may not be as rich, it’s not as responsive. And of course, the performance will vary. I think what’s likely to happen is, there’s going to be some amount of buyer’s remorse. Users are going to feel like they gave up a lot of rich features and functionality. It’s not the same, experiencing an application through a browser. It’s not going to come close.”

The problem with the browser being the front-end for SaaS applications is that users have very clear expectations of a browser experience, based on their use of the Web. Users go to a site and expect it to load fast — in two seconds or less. They don’t like to wait, and won’t. Google says that for every additional 500ms of delay, the site loses 20 percent of its traffic. With fast broadband and wireless connections everywhere, users expect blazing speed when they fire up their browser.

“You need to understand the browser,” says Robert Hughes, director of Global Services & Solution Consulting at Keynote. “Browsers are now full-fledged application environments, rendering, executing, initializing, and transacting every day.”
Even though there are a whole string of opportunities for performance to degrade between the SaaS data center and the enterprise user, what happens in the browser may have the biggest impact.

“Researchers have found that 75 percent of the performance problems with Web applications actually happen in the browser,” Agarwal says. “Not in the data center, not in the database, not in the network or anywhere in the back end, but in the browser. And so the way you architect the user interface with Javascript and all the other browser technologies is really what is going to impact the performance of the service.”

And then there’s the 25 percent of opportunities for something to go wrong that are not browser-related — CDNs, domain name servers, routers, load balancers, client-side scripts, browser add-ons, etc. Understanding where along the line performance is compromised is the key to delivering applications that will satisfy and retain users.

Measure Your Way To Success

If performance is the key to success, measurement is the key to performance. For both SaaS vendors and enterprise customers, the combination of active performance monitoring and robust, enforceable service level agreements is essential for successful, profitable operations.

SaaS vendors will keep their customers happy with proactive performance management and, if written into SLAs as standard operating procedure, will have a distinct marketing advantage over those who do not. Additionally, a clear understanding of where and when slowdowns occur can help keep their partners honest — if the bottleneck is with one of their vendors — and also protect and maximize their return on investment. For example, investment in additional servers or network capacity may not be required, if the problem can be pinpointed to the application-browser interaction, a third-party component, or at some other point along the path. Performance visibility enables SaaS vendors to invest their resources where they will reap the highest return. And to get that kind of visibility requires stepping outside the walls of the data center.

“You have to understand how it’s experienced as an end user,” Agarwal says. “You have to monitor the true experience, look at the transaction and look at where the slowdown is. You have to see whether it’s a data center problem, whether it’s the network, or whether it’s just the way you built the application that is now being accessed by your customers through their browsers.”

“Your users are not sitting in a data center,” Hughes adds. “Test from the Internet, test with a browser, test with what your users use.”

Turning Incredible Promise Into Credible Reality

New technology adoption is always a push-pull dance. Enterprises want the productivity and competitive advantage, but are understandably cautious until the risks are understood and performance proven. New technology vendors want to get out ahead of their competitors and gain preemptive market share quickly, sometimes before all the kinks are worked out. For both sides, it boils down to credibility.

Some SaaS vendors are proactively putting their performance stake in the ground and backing it up with guarantees. RightNow, a CRM application provider, starts giving money back to clients if up-time drops below 99.9 percent. Intacct, which offers financial management and accounting applications, offers a similar guarantee, and publishes their uptime stats right on their Web site.

“Web performance monitoring helps you to protect your bottom line, but also to build your brand and build customer loyalty,” says Neeraja Rasmussen, Keynote senior marketing manager for Web Performance, about SaaS vendors. “You can guarantee that you can deliver a certain level of service. Verified third-party data with high levels of trust is a definite competitive advantage for SaaS providers.”
Such data is the heart of effective, enforceable SLAs. And in a nascent SaaS marketplace, solid SLAs are imperative for credibility.

A Key Component Of The Connected Future

The lines are increasingly blurred between what happens on our computers, and our phones, and our TVs, and what happens in the cloud. What’s important is getting the tasks done efficiently and effectively. Saas will no doubt play a bigger and bigger role in accomplishing that end, though we eventually may stop making that distinction. It will all just be “computing.”

Back to Top