Discover PerformanceHP Software's community for IT leaders // June 2012
SaaS without silos
Software as a service offers speed, agility and cost efficiency—all good. But if those solutions create inefficient silos of data? Not so good.
Enterprise IT is undergoing a transformation. (Isn’t it always?) This transformation includes a shift from on-premise sourcing of every application solution. Software as a service offers flexibility, speed and savings, while letting IT focus internal resources where they’re most needed. This has allowed IT to get creative about acquiring not only applications for the business, but also the very lifecycle management capabilities needed to create, modify and support those apps.
This shift to SaaS-based lifecycle management tools is a good thing. It helps to put the best solutions into developers’ hands as quickly as possible, and at the lowest cost. However, this requires solid management of SaaS procurement and deployment. Otherwise, that easy, inexpensive SaaS option can isolate teams and minimize visibility. Too much ad hoc procurement could return delivery organizations to a world of point-tool islands and siloed data environments—when IT spent a half-decade on application lifecycle management to escape exactly that scenario.
Apps leaders should embrace SaaS for its virtues—but before choosing which solutions to deploy, ask yourself three questions.
First question: What’s going on now?
The first order of business: assessing your current state. How much SaaS is used in your organization today? Of that, how much has been deployed without IT’s involvement? Try to generally characterize the state of SaaS in your organization. Have your current SaaS solutions created new data silos? Are SaaS benefits being undercut by poor management?
Also, what are your official policies around SaaS use and provisioning? If you’re like many organizations, use of SaaS has grown so fast and is so decentralized that you don’t have any policies. That might have been fine before, but given the continuing growth of SaaS, you’ll soon be dealing with an explosion of credit card provisioning and resulting silos you aren’t even aware of.
Second question: What’s your ideal situation?
Even if SaaS is already sprawling throughout your organization, you can’t rein it in until you know how it should be treated as part of overall IT planning. What kind of apps do you envision “outsourcing” to SaaS? You might be OK with external provisioning of anything that’s not a core app, or you might limit SaaS to quick, cheap solutions for developers. Create a list of criteria for determining what is core and non-core, and therefore what should be SaaS sourced and what shouldn’t.
Third question: How will you evaluate potential SaaS solutions?
Your evaluation may include qualities that are uniquely important to your organization, but here’s a good start:
- Ability to share. Will this solution silo your data, or is it open and easy to integrate? If it creates a silo, is that acceptable for this particular type of data?
- Ability to grow. Does the offering grow as your needs grow? As you evaluate SaaS vendors, look for capabilities to support the entire lifecycle. A solution may solve your immediate need, but you might need new tools 12 months down the road. The vendor should have a next-level offering so you can migrate from your initial investment to the next capability.
- Security. Does the solution meet your organization’s security requirements for SaaS provisioned solutions—which may differ from on-premise requirements?
- Lifecycle. What happens to your data when you retire this system? Can you retain it and migrate it as necessary?
- Ops needs. Eventually, you’ll need to create that official policy around SaaS provisioning, and that would be a good time to engage Ops. What are the Ops team’s needs around supporting SaaS solutions? Can this solution support those needs?
A little thought goes a long way
The proliferation of SaaS presents a tremendous opportunity for Apps teams. It provides effective solutions in a fraction of the time it would take to build them yourself. But it must be managed to avoid creating issues of siloed data and application redundancy. When investigating ALM solutions, prioritize those that unify teams and data, regardless of whether on-premise or SaaS-based.
Read about HP’s Application Lifecycle Management solution via SaaS.
Register for HP’s premier event for inspiration from industry leaders, the HP inside scoop, and a deep dive into tomorrow’s enterprise IT trends.
HP Software VP Paul Muller brings in HP and industry-wide experts each week for challenging discussions about trends in Big Data, mobility, IT security and more.
Forrester analyst Kurt Bittner discusses how to enhance Agile development with continuous integration and automated testing to deliver real business results.