Killer Apps

Is your applications portfolio is a hodge-podge of programs developed over the past 20 years, many of which require life support? Is your e-business strategy driven by a mix of mainframe-based, client-server, and Internet applications? Do you support some in-house, outsource support for others? Time for an applications health check-up.
There's a good chance that your applications portfolio is not what it should be. It's probably got a hodge-podge of applications developed over the past 20 years or so, many of which require some form of life support to exist. Your e-business strategy is probably driven by a mix of mainframe-based applications, client-server applications, and Internet applications. You support some of them in-house and outsource support for others, most likely your e-business applications.

Time for an applications health check-up.

Your e-business requirements will define an enterprise applications strategy that should marry back-office applications with front- and virtual-office ones. You'll need to connect your employees with your customers and suppliers. You'll also need to retain enough flexibility in your environment to accommodate enhancements and whole new strategies, even as you juggle mobility requirements.

Your goal is a set of inter-related, interoperable applications for back-office, front-office, virtual-office, desktop, personal digital assistant, and other thin client applications. All must support your business strategy. These applications should be standardized and should support activities, processes, employees, customers, and suppliers, regardless of where they physically sit or how mobile they are. You should review all of your applications to determine their compatibility with this goal.

Elements of Applications Alignment

Let's look at nine elements of an applications alignment strategy:

1. Business strategy linkages
Make sure you understand what the business wants to accomplish with the applications you buy, build, and support. If you can't put your finger on "purpose," then your applications portfolio will, by definition, perform less than optimally. You should deploy all applications based on their measurable contribution to business results.

2. Applications portfolio assessment
Look at your applications objectively. Which contribute measurably to your business? Which require disproportionate support? It's essential to assess applications with reference to your business strategy and the relative contribution they're making to business processes and profitability. If the outcome of that assessment is clear, then you need to make some decisions. You'll need to decommission expensive applications that contribute little, or, for older applications that are limited but still valuable, transfer functionality to systems less expensive to maintain.

Assess the variation in your applications portfolio. How many architectures are you supporting? What's the distribution of functionality and architecture type? Do you have your most important applications on the oldest, most-expensive-to-maintain platforms?

You need a standard applications architecture. If you keep buying and integrating different host-based, client-server, Internet, and hybrid architectures, your support costs will rise as rapidly as your reliability declines.

You need to think about the range of applications in your portfolio, how you procure them, how you support them, and how you replace or modernize them. A framework that might suit your purposes is shown in the table below.


			Corporate		        Personal

PDAs/		        E-mail & groupware		E-mail & calendaring
Thin Clients		Calendaring			Personal transactions
		        Browsers			Instant messaging


Desktop/		Financial management 	        Word processing		
Laptop		        Communications 		        Presentation graphics		
			Knowledge management 	        Browsers			

Enterprise		Legacy				Training & education
			Packaged ERP			Productivity tools
			Internet, intranet		Knowledge management

Management		Network & systems		Project management
			management			Program management
			Applications management	        Workflow

Services		End-to-end services		Information management
			Horizontal services		Searching
			Vertical services		Configuration

Enabling Technologies    Middleware			Voice recognition
			Artificial intelligence		Fingerprint recognition
			Components			Artificial intelligence

Figure 1: An applications assessment and planning framework

3. Desktop/laptop/thin applications
Stay as standard as possible here. Go with the mainstream vendor, Microsoft, and stay within primary standards with non-Microsoft applications. The primary desktop and laptop applications include the office suite running on a Windows platform. While some of us may find this ubiquity disconcerting, it makes the world compatible.

Watch the upgrade treadmill. Upgrading hundreds or thousands of desktops to get access to a few cool features makes no sense at all. Skip versions whenever possible.

PDA applications are still evolving. We're a couple of years away from serious skinny applications that run on robust handheld operating systems. Until then, synchronization will provide integration with a suite of desktop and laptop messaging and calendaring applications.

4. Enterprise applications
Decisions here can make or break your company. Does your organization need to get out from under myriad home-grown applications? Will your culture sustain a long implementation? Can your business adapt to the business models embedded in enterprise resource planning (ERP) applications? Is the value of standardizing on processes, reporting, and management very high?

If the answer to all these is yes, consider an ERP application. But be careful. If any part of the organization bails on the effort, the whole thrust will crash and burn. Migrating to an ERP application must also provide necessary impetus to your e-business strategy. In other words, a project of such momentous proportions must look forward, not backwards.

5. Applications management
Beyond ERP applications are network and systems management enterprise applications that can also add value -- or pain -- to your organization. The same decision criteria listed for No. 4 apply here. In addition, the sheer size of your organization has to justify a robust management suite. Regardless, to support mainframe applications you'll need to manage your distributed applications with different tools and processes than you're now using.

Other enterprise management options include enterprise-wide project and program management training. Strongly consider instituting programs to get your staff up to speed. Enterprise-wide training is also a good idea, especially when your organization suffers significant skill-set gaps.

6. Applications services
Evaluate your ability to design, develop, integrate, support, and modify your applications. If you have a large, well-trained internal staff capable of supporting all varieties of applications and architectures, then you might want to stay where you are. But if you already have many applications and are moving faster and faster toward distributed e-business applications, then you should strongly consider outsourcing.

In short, you now need better reasons NOT to outsource than to keep it all in-house -- the direct opposite of decision-making processes popular five years ago.

7. Enabling technologies
Keep an eye on enabling technologies most likely to affect your applications performance and costs. Access devices, such as biometric fingerprint authentication tools, can enable single-sign-on and reduce security administration costs.

Other technologies, like software component technologies, can help your support and modification processes. They should be considered.

Other human-computer interface (HCI) technologies should be monitored as well, since speech input/output, for example, will revolutionize applications performance.

Middleware will continue to be a critical applications technology. Don't leave it to the geeks! As the glue that makes the applications work together, it's important that you understand your overall middleware strategy.

Finally, artificial intelligence will remain hot for some time to come. Track progress here closely, especially as it's used to beef up applications monitoring and performance, especially automated transaction processing.

8. Future modeling
Institutionalize a process that reviews at least twice a year the whole applications strategy, driven (ideally) by new business models. Track key applications technology to determine what your migration plans might look like.

9. Ownership and communication
Finally, you need to own the applications technology strategy, which, in turn, should be driven by a jointly-owned business strategy.

Communications need to be clear and couched in meaningful costs and benefits. For instance, what it will cost to implement and support recommended applications technology -- and what will lost opportunities cost if the technologies are not implemented? And what is the value of the business benefits -- increased customer satisfaction, retention, cross-selling, and the like?

Steve Andriole is the founder & CTO of TechVestCo, a consortium that focuses on optimizing IT investments. He is a former chief technology officer of both Safeguard Scientifics and CIGNA Corp. His career began at the Defense Advanced Research Projects Agency, where he directed cybernetics technology.






0 Comments (click to add your comment)
Comment and Contribute

 


(Maximum characters: 1200). You have characters left.